design, procurement, and construction … · ... project management doc. no.: p-rpt-j-00031 ......

SALT WASTE PROCESSING FACILITY

DESIGN, PROCUREMENT, AND CONSTRUCTION

LESSONS LEARNED AND BEST PRACTICES

Contract No. DE-AC09-02SR22210

Function: Project Management

Doc. No.: P-RPT-J-00031

Revision: 0

Date: 02/10/2017

SWPF Design, Procurement, and Construction Lessons Learned and Best Practices P-RPT-J-00031, Rev. 0

Frank Sheppard, Jr.

SWPF Project Manager

SIGNATURE PAGE

Signature Page

Date

SWPF Design, Procurement, and Construction Lessons Learned and

Best Practices P-RPT-J-00031, Rev. 0

Summary of Changes

SUMMARY OF CHANGES

Revision No. Date Description of Change

A 12/05/2016 Draft Issued to DOE.

A 12/12/2016 Draft

A 12/20/2016 Draft

A1 02/01/2017 Develop per DMR-3750. Issued for Intradiscipline Check.

A2 02/02/2017 Develop per DMR-3750. Issued for Interdiscipline Review.

0 02/10/2017 Develop per DMR-3750. Issued as Approved.



Table of Contents

TABLE OF CONTENTS

1.0 INTRODUCTION..............................................................................................................1

2.0 BACKGROUND ................................................................................................................1

2.1 Physical, Chemical, Radiological, and Technological Framework ...................1 2.2 Legislative, Regulatory, Stakeholder and Oversight Framework and

Impacts ....................................................................................................................3

3.0 LESSONS LEARNED AND BEST PRACTICES ..........................................................4

3.1 Management ...........................................................................................................5 3.1.1 DOE - Contractor Relations .....................................................................5 3.1.2 Management System Development and Planning ...................................6 3.1.3 Nuclear Design-Build Experience .............................................................7 3.1.4 Unrealistic Milestones: Managing Cost and Schedule Risks .................8

3.1.5 Data Management ....................................................................................10 3.2 Contracts and Procurement ................................................................................13

3.2.1 Capital Asset Acquisition Strategy .........................................................13 3.2.2 Procurement .............................................................................................17

3.3 Design ....................................................................................................................21

3.3.1 Design Authorization Basis .....................................................................22 3.3.2 Integrating Design Engineering and Facility Safety .............................23 3.3.3 Maximize Design Margin ........................................................................25

3.3.4 Full Scale Design Testing ........................................................................25

3.3.5 Engineering Team Interface with Stakeholder .....................................26 3.3.6 Balancing Design Perfection and Project Progress ...............................26

3.4 Construction .........................................................................................................27

3.4.1 Craft Recruitment and Retention...........................................................27 3.4.2 EPC Contract Impact on Construction Performance ..........................28 3.4.3 Collocation of Construction and Engineering Staff ..............................29 3.4.4 Constructability Review ..........................................................................29

3.5 Transition from Construction to Testing ...........................................................30

4.0 REFERENCES .................................................................................................................33

List of Appendices

Appendix A. SWPF Project Cost Data for Design and Construction Appendix B. Data Management Approach, Lessons Learned and Best Practices.



Acronyms and Abbreviations

LIST OF ACRONYMS AND ABBREVIATIONS

A&E Architecture and Engineering

ANSI American National Standards Institute

ASME American Society of Mechanical Engineers

ASP Alpha Strike Process

BOBCalixC6® Calix[4]arene-bis(tert-octylbenzo-crown-6)

CD Critical Decision

CGD Commercial Grade Dedication

COR Codes of Record

CRFI Construction Requests for Information

Cs Cesium

Cs+ Cesium Ions

CSE Cognizant System Engineer

CSSX Caustic-side Solvent Extraction

DBB Design-Bid-Build

DCD Design Criteria Database

DCN Design Change Notice

DMCS Document and Material Control System

DNFSB Defense Nuclear Facilities Safety Board

DOE U.S. Department of Energy

DSA Documented Safety Analysis

DWPF Defense Waste Processing Facility

EDMS Engineering Document/Data Management System

EPC Engineering, Procurement, and Construction

ES&H Environmental, Safety, and Health

FHA Fire Hazard Analysis

HAZOP Hazard and Operability

HVAC Heating, Ventilating, and Air Conditioning

IT Information Technology

ITP In-Tank Precipitation

K+ Potassium Ion

MST Monosodium Titanate

Na Sodium

Na+ Sodium ion

NaOH Sodium Hydroxide

NCR Nonconformance Report

NDAA National Defense Authorization Act

NPH Natural Phenomena Hazard

NRC Nuclear Regulatory Commission

NWPA Nuclear Waste Policy Act of 1982

P&ID Piping and Instrumentation Diagram

PDSA Preliminary Documented Safety Analysis

PHA Preliminary Hazard Analysis

PLA Project Labor Agreement



Acronyms and Abbreviations

LIST OF ACRONYMS AND ABBREVIATIONS (cont.)

PO Purchase Orders

QA Quality Assurance

QAP Quality Assurance Program

QC Quality Control

S/RID Standards/Requirements Identification Document

SDR Supplier Deviation Request

SME Subject Matter Expert

SOT System Operational Test

SQA Software Quality Assurance

SQL Structured Query Language

SR Strontium

SRS Savannah River Site

SSC Structure, System, and Component

SWPF Salt Waste Processing Facility

TPB Tetraphenylborate

UAI Use-As-Is

WIR Waste Incidental to Reprocessing



Page 1 of 35

1.0 INTRODUCTION

The U.S. Department of Energy (DOE) Savannah River Site (SRS) and its prime contractor

Parsons declared construction complete for the Salt Waste Processing Facility (SWPF) on April

22, 2016. SWPF is designed and constructed to safely treat millions of gallons of liquid

radioactive waste, presently stored in underground tanks at the SRS F- and H-Area Tank Farms.

This Lessons Learned and Best Practices review was developed at the request of the DOE SWPF

Federal Project Director. This report describes major risk mitigation tactics and strategies that

should be considered for future nuclear facility acquisitions.

Parsons declared Construction Complete in April 2016, eight months ahead of schedule and

approximately $68 million under budget against Contract Modification 116 of the SWPF

Contract (DE-AC09-02SR22210, Design, Construction, and Commissioning of a Salt Waste

Processing Facility [SWPF]1) (See Appendix A). DOE accepted the declaration in May 2016.

Total Project Cost to complete Testing, Cold Commissioning, Startup, and One Year of

Operation is estimated to be $2.3 billion with a Project completion date of February 2021. The

projected Total Project Cost is below the original Critical Decision-0 upper bound June 2001

$2.6 Billion estimate for the completion of the SWPF Project. The Project is currently in the

testing phase of Plant Commissioning. This will entail calibration, grooming, and alignment and

detailed testing of 60 systems. Sixty System Operational Test (SOT) and five Integrated SOT

procedures have been completed and approved. Parsons continues to work constructively with

DOE to successfully bring this key risk reduction facility online to facilitate the clean-up mission

of the SRS. Parsons and DOE are striving to accelerate preparations for post-construction testing

and commissioning activities to maximize the probability that start-up of the facility is achieved

on or ahead of schedule. Parsons and DOE are engaged with the site Liquid Waste Operations

Contractor to ensure that requisite interface integration and support capabilities will be in place.

This will ensure that the SWPF will receive necessary feed and have adequate effluent receipt

facilities.

2.0 BACKGROUND

This section describes the technological and regulatory context of the challenges associated with

treating and disposing of SRS F- and H-Tank Farm radioactive waste. The technical challenges

required decades of research to overcome. The regulatory issues emerged early in the Project’s

design phase. Both technological and regulatory hurdles were overcome with a clear path to

completing the salt waste treatment and disposal mission.

2.1 Physical, Chemical, Radiological, and Technological Framework

Nuclear material production at the SRS generated millions of gallons of radioactive waste.

Approximately 37 million gallons of this waste is stored in 49 existing underground storage

tanks in the F- and H-Area Tank Farms. The waste in the tanks is in three forms:

1. Three (3) million gallons of Sludge composed of metal oxides making up one-half the total

inventory of radioactive material in the tanks;



Page 2 of 35

2. Sixteen (16) million gallons of salt precipitate or saltcake; and

3. Eighteen (18) million gallons of salt solution supernate.

Including water needed to dissolve the salt plus ongoing waste streams, the salt solution volume

that needs to be processed is approximately 100 million gallons.

WI-TR-94-0608, Approved Site Treatment Plan2 and WSRC-OS-94-92, Federal Facility

Agreement for the Savannah River Site3 require glass immobilization of radioactive tank waste.

Waste is vitrified in stainless steel canisters at that Defense Waste Processing Facility (DWPF).

The immobilized waste will be transferred to the National Geological Repository pursuant to

Nuclear Waste Policy Act of 19824 (NWPA). SWPF’s mission is to reduce the vitrified waste

volume to make glass immobilization economically feasible. The Waste volume requiring

processing at DWPF and disposal at the repository will be significantly reduced by separating

Cesium (Cs), Strontium (Sr), and Actinides from the salt waste, generating a concentrate for

immobilization in glass and a decontaminated waste stream for grout immobilization at the SRS

Saltstone Facility.

Fabricating large underground storage tanks from stainless steel was impracticable in the 1950’s

and early 1960’s. The tanks were fabricated from carbon steel. Carbon steel is incompatible with

even dilute acid. To achieve compatibility, the pH of the acidic waste stream from F- and H-

Canyon was adjusted with concentrated Sodium Hydroxide (NaOH). This achieved compatibility

but rendered separation of Cs from the NaOH difficult. Sodium Ions (Na+) and Potassium Ions

(K+) behave identically to Cesium Ions (Cs+) in simple precipitation or ion exchange reactions.

Separation required more complex compounds designed to have a stronger affinity for

partitioning Cs+ than Na

+ and K

+. Tetraphenylborate (TPB) was identified in the 1980s as having

this capability. Mixing Na-TPB with waste results in a Cs-TPB precipitate that can be separated

and vitrified. Full scale TPB processing started in 1995 using In-Tank Precipitation (ITP). ITP

operations were suspended after higher than expected concentrations of benzene were generated

from the breakdown of TPB by radiolysis and catalytic decomposition processes (Defense

Nuclear Facilities Safety Board [DNFSB] Recommendation 96-15). Due to benzene’s high vapor

pressure and low flammability limit, operations were abandoned in1999 due to safety concerns.

The DOE responded by identifying three alternative technologies. DOE/EIS-0082-S2, Savannah

River Site Salt Processing Alternatives Final Supplemental Environmental Impact Statement.6

and Record of Decision: Savannah River Site Salt Processing Alternatives. U.S. Department of

Energy, Washington, D.C. Published in the Federal Register: October 17, 2001 (Volume 66,

Number 201)7 documented DOE's analysis and selection of SWPF technologies. The preferred

technologies included the Monosodium Titanate (MST)-based Alpha Strike Process (ASP) for

removal of actinides and Sr, and Caustic-side Solvent Extraction (CSSX) for the removal of Cs.

CSSX, developed at Oak Ridge National Laboratory, is based on a crown ether molecule,

designed to have a very high affinity for Cs+ relative to Na

+ and K

+. The CSSX solvent is a

solution of Isopar®

L (highly purified kerosene) with 0.007 moles per liter Calix[4]arene-bis(tert-

octylbenzo-crown-6) (BOBCalixC6®) with other trace compounds to enhance stability and

longevity. CSSX is a continuous process that utilizes centrifugal contactors for Cs extraction.

The Cs is removed by contacting aqueous salt solution with the organic solvent in a series of



Page 3 of 35

extraction stage contactors. Cs is captured in the organic solvent and then removed from the

solvent by a series of stripping contactors that mix the solvent with a slightly acidic strip

solution. The strip effluent is transfer to the DWPF for vitrification. The decontaminated salt

solution is transferred to the Saltstone Facility.

2.2 Legislative, Regulatory, Stakeholder and Oversight Framework and Impacts

DOE approved Critical Decision (CD)-0, Mission Need, on June 25, 2001. Phase I of the Project

began in September 2002 as a competitive bid between two teams, down-selected from an initial

group of bidders. Each team prepared a conceptual design that included a Preliminary Hazard

Analysis (PHA), Basis of Design, Design Criteria Database (DCD), Functional Specification,

Feed Specification, Standards/Requirements Identification Document (S/RID), and

Environmental Plan. The design and associated analyses were used to develop a cost and

schedule proposal for Preliminary and Final Design, Construction, Testing and Commissioning,

and one year of Operations. The Parsons team was awarded Phase II of the Contract in January

2004 based on the Conceptual Design and bid package.

Phase II design was delayed several times due to legal, regulatory, and stakeholder issues. The

first delay was directed in March 2004 by the DOE Assistant Secretary of Environmental

Management due to a Federal court ruling implicitly prohibiting SWPF operations. The U.S.

District Court of Idaho ruled in July 2003 (National Resource Defense Council versus Abraham

20038) that NWPA

4 requires disposal of High-Level waste at the Nuclear Regulatory

Commission (NRC) Licensed National Geological Repository. This ruling would require all

waste, decontaminated or not, to be vitrified, making the SWPF process unnecessary. The US

Court of Appeals 9th

Circuit vacated the lower court ruling in November 2004 (National

Resource Defense Council versus Abraham 20049) since the lawsuit and ruling pertained to a

future and therefore nonexistent condition. The case highlighted the need for a risk-based

alternative approach to the NWPA4 requirement. NWPA

4 provides no alternative treatment and

disposal standards for decontaminated waste, regardless of risk/waste composition. To address

this shortcoming of the NWPA4, Congress included a provision in the National Defense

Authorization Act10

(NDAA) that provides for consideration of waste composition and risk.

Section 3116 of NDAA10

granted authority to the Secretary of Energy, in consultation with the

NRC, to reclassify decontaminated salt solution as Waste Incidental to Reprocessing (WIR) at

DOE facilities in South Carolina and Idaho. WIR is not regulated under the NWPA4 and can be

disposed in accordance with NRC radioactive waste performance standards. The President

signed the NDAA10

in October 2004. DOE Approved CD-1 on August 12, 2004 with

Congressional and Executive Office assurances of approval. The NWPA4 issue resulted in a six

month delay in the start of Preliminary Design.

As the WIR issue was resolved, issues concerning the seismic performance standard were raised

by the DNFSB. Although not an existential threat to SWPF and the SRS liquid waste treatment

strategy, it ultimately resulted in a two year delay in schedule. Conceptual Design and the PHA

were provided to the DNFSB after the Phase II contract award. Consistent with DOE standards

stipulated in the Contract (DE-AC09-02SR222101) and the PHA, SWPF was designated as a

Performance Category (PC)-2 facility in the Conceptual Design. The DNFSB issued a letter to



Page 4 of 35

the Deputy Secretary of Energy on August 27, 2004, stating that based on recent DOE guidance,

SWPF should be designed as a PC- 3 facility. Ensuing discussions of opposing positions over the

appropriate application of PC-3 continued for 14 months while Preliminary Design of a PC-2

facility progressed. DOE accepted the DNFSB recommendation on November 22, 2005,

directing Parsons to redesign the facility to meet seismic PC-3 requirements. This change

translated into a delay in Final Design. The standards in the DOE-approved Codes of Record

(COR) and PHA unambiguously required a PC-2 design. Opposing positions were no doubt

intended in the best interests of the public. The point is not to question the decision; however, the

circumstances associated with the decision’s timing affords a valuable cautionary message for

future projects. These issues should be resolved before Preliminary Design.

The DNFSB continued to raise concerns over the geotechnical investigation and the structural

design’s capacity to meet PC-3 standards. Enhanced Final Design addressed these concerns by

increasing the thickness of the base mat. Enhanced Final Design completion was announced in

December 2008. CD-3, Start of Construction, was approved on January 19, 2009 by the Deputy

Secretary. The approved Total Project Cost increased from $900 million to $1,330 million and

extended CD-4, Project Completion from November 2013 to October 2015. Delays thereafter

were due primarily to supplier quality problems and late deliveries of equipment and materials.

3.0 LESSONS LEARNED AND BEST PRACTICES

The regulatory and technical challenges that faced SWPF during design were an expected part of

the nuclear design-build process. If a project survives the initial challenges, ongoing

contingencies from stakeholders, regulators, and a lack of nuclear expertise, and suppliers will

continue to challenge completion. To effectively respond to ongoing challenges, leadership must

establish a culture where:

1. It is understood that safety and quality are essential to controlling cost and schedule;

2. Technical inquisitiveness that challenges assumptions is encouraged and cultivated;

3. Dedication to excellence and expertise takes precedence over turf and the status quo;

4. Procedure compliance is balanced with equal emphasis on stopping if the procedure is not

understood, unsafe, or adverse to quality; and

5. Respect, professionalism, and open communications are fostered at all levels. This includes

the relationship between the Contractor and DOE, and external oversight organizations.

These characteristics are necessary conditions for effective communication, the only antidote to

the complexity of a nuclear design-build project. An organization dominated by group think, a

focus on cost, a permissive attitude toward procedure compliance, or mistrust and disrespect

between organizations and regulators, will find it very difficult to maintain necessary safety,

quality, and productivity performance standards. Poor integration and communication between

contributing organizations will result in errors that propagate across design outputs,

procurements, and installed Structures, Systems, and Components (SSCs).



Page 5 of 35

The Project’s successes were achieved by adapting to internal and external challenges. This is

never easy or straightforward. Improvements and corrections commonly have unanticipated

consequences, requiring further refinements and adjustments. Becoming a learning culture

requires discipline and experienced leadership.

The following describes and discusses conditions essential to successful completion of a nuclear

design-build Project. If the past is any indication of the future, then success will not come

without challenges that impact both cost and schedule. The most significant challenges were the

sources of the most important lessons learned. The best practices were commonly part of the

longer term adjustments, leading to the Project’s completion. Key best practices or critical

success factors included:

A team with nuclear experienced leadership, management, and staff;

DOE’s selection of a Major System Acquisition strategy that was based on an Engineering,

Procurement, and Construction (EPC) Contract;

The establishment of an effective integrated data management program early in the design

process; and

Quarterly partnering meetings between DOE and Parsons which provided an effective means

of managing owner-contractor relations. These meetings were established as a lessons

learned to strained relations that emerged during schedule delays caused by late delivery of

the Large ASME Vessels.

The major lessons learned included the following:

Suppliers for fabrication of critical, high-risk engineered items should be selected based on

Best Value versus Lowest Price, Technically Acceptable criteria;

A fixed set of fundamental design requirements should be agreed upon at CD-1 and remain

unchanged throughout Preliminary and Final design; and

Realistic milestones should be established at CD-1 that are based on well developed

assumptions with adequate management reserve and contingency.

These factors do not exist in isolation. Without experienced leadership, management, and staff,

the full importance of these factors are imperfectly understood, leading to unrealistic milestones

and major quality issues.

3.1 Management

3.1.1 DOE - Contractor Relations

Project success is strongly dependent on a relationship based on trust between the DOE and

contractor. A dysfunctional relationship characterized by mistrust results in inefficiencies caused

by poor communications and second guessing decisions. More significantly, a dysfunctional

relationship is a strong signal of a failing Project. Capital asset project’s like SWPF are under

close scrutiny by DOE-Headquarters; Congress; state governments and their regulatory agencies;



Page 6 of 35

oversight organizations such as the DNFSB; and nongovernmental stakeholders. In an

environment of diminishing discretionary budgets, a capital asset project that appear to be failing

or in trouble is likely to jeopardize funding and support from Congress.

The Project has experienced both healthy and dysfunctional DOE-contractor relations. The first

years of the Project were characterized by an effective and trustful relationship. As issues with

cost increases and supplier delays emerged, insufficient attention was placed on maintaining a

generative relationship. This dysfunction was remedied by changing DOE and Contractor Project

leadership and the establishment of quarterly Partnering Sessions. The initial meetings were

mediated by a third party and were characterized as tense. Continued efforts from leadership in

both organizations was successful in building trust through honest and candid dialogue. As a

lessons learned and best practice, Projects should institutionalize regularly scheduled partnering

sessions between the two organization’s leadership teams. In addition to the partnering sessions,

it is a best practice for the respective disciplines to hold weekly meetings to resolve issues (e.g.,

Engineering, Quality Assurance [QA], and Commissioning).

3.1.2 Management System Development and Planning

The Nuclear Regulatory Commission’s lessons learned report to Congress on failed commercial

nuclear power plant design-build projects found that quality issues which were the direct cause

of failure, were largely attributed to broad management system failures (NUREG-1055,

Improving Quality and the Assurance of Quality in the Design and Construction of Nuclear

Power Plants11

). The management system depends on the quality of the policies, plans,

procedures, training, and personnel. It is common for Projects to underestimate the cost and time

required to establish the required system of directional documents and training.

A competitive business environment among design-build companies has prevailed for decades,

driving low overhead rates that preclude the retention of nuclear-experienced staff and

management systems that are necessary for nuclear projects. Delivery of a completed nuclear

facility includes both the physical plant and a substantial set of design conformance verification

records (> 500,000 for SWPF). The rigor and complexity of nuclear design-build projects require

a sophisticated, complex array of policies, programmatic plans, administrative procedures, and a

means of training personnel. The number of design-build companies, fabricators, and suppliers

that maintain the necessary set of written programs and procedures is constrained by sporadic

demand. An off-the-self, documented nuclear management system is unlikely available at a

corporate level. Necessary programs and procedures must be developed as the work progresses.

This requires nuclear-experienced managers, supervisors, engineers, regulatory compliance

experts, and nuclear safety experts at the beginning of the project. The population of personnel,

who understand the unique requirements, and necessary programs and processes, especially as it

applies to understanding the necessary documentation, is very small. Early recruitment of

Subject Matter Experts (SMEs) in nuclear-specific fields is a critical success factor and a lesson

learned throughout this report.

Between the early 1970’s and the late 1970’s, the cost of nuclear power plant design-build

projects increased by an order of magnitude. This increase was due to the cost of complying with



Page 7 of 35

Nuclear Regulatory Commission regulations, particularly those associated with QA. The

increased cost was for the additional “hotel load” of SMEs, clerical staff, supervisors, and

manager to oversee these personnel (Cohen 1990, The Nuclear Energy Option12

). Government

decision makers, stakeholders, the media, and management consultants that lack nuclear design-

build experience, understandably fail to grasp the need for funding profiles with level-of-effort

costs that approach or exceed the discrete costs. The Project was under constant pressure to

reduce the “hotel load.” The Project reduced projected funding profiles for quality and

procurement staff during Preliminary Design only to hire more than planned to recover from

issues with substandard suppliers. The need to have sufficient nuclear experience personnel in

Conceptual and early Preliminary Design to develop the required management system processes

is a recurring lessons learned in this report. This includes Software QA (SQA), Technical

Representative (Procurement Engineers), and QA, Quality Control, (QC), and Document Control

SMEs.

3.1.3 Nuclear Design-Build Experience

The nuclear design-build experience of the leadership, management, and staff was essential for

successful completion of the SWPF. This best practice is simple and straightforward; however,

the underlying reasons are commonly given cursory treatment by decision makers that do not

have nuclear design-build experience. This report develops and describes the unique nature of

nuclear design-build risks as they pertain to management, Procurement, Contracting,

Engineering and transition to Commissioning. This section summarizes the major challenges that

require experience to effectively respond to challenges to maintain project progress.

All complex design-build projects, nuclear or not, are a relentless series of emerging

contingencies and counter responses. Success is predicated on having experienced management

and staff that are familiar with these contingencies and can recognize and respond quickly with

the appropriate counter measures. The report will explain why the number and significance of

risks increases for nuclear design-build projects. DOE nuclear design-build projects have another

degree of challenge derived from its unique missions. SWPF and its predecessor ITP are first-of-

a-kind facilities, designed and constructed to address one of DOE’s unique missions. DOE

nuclear facilities are commonly the prototype and final model. Prototypes are built to identify

risks and either 1) develop the necessary controls to safety go into full production or 2)

determine from the prototype that the approach is infeasible. ITP was essentially a prototype

determined to have uncontrollable risks. Full-scale testing mitigates risk, but it is not a panacea.

ITP was tested and studied for a decade before construction. These projects demand leadership

and staff who have faced the unique contingencies associated with first-of-a-kind, nuclear

design-build projects and who have the technical expertise and judgment to make the right

choices as contingencies arise.

The importance of nuclear leadership and management experience was recognized in NUREG-

105511

. The study found that troubled projects were led by corporate and or project management

who lacked prior nuclear experience and whose previous experience was confined to fossil

power plants. The latter had the negative impact of imparting a false sense of expertise. NUREG-

105511

concluded those utilities that struggled:



Page 8 of 35

“… approached their nuclear projects as extensions of the earlier fossil construction

activity, i.e., to be managed, staffed, and contracted out in much the same way as fossil

projects. The utilities did not fully appreciate or understand the differences in complexity,

quality requirements, and regulations between fossil and nuclear projects . . . lack of

experience in and understanding of nuclear construction manifested itself in . . .

inadequate staffing, selecting contractors … who had limited nuclear experience; …use

of contracts that emphasized cost and schedule; … lack of management support for

quality programs; … lack of appreciation of the ASME codes and other nuclear

standards; . . . inability to recognize that recurring problems were merely symptoms of

much deeper, underlying programmatic deficiencies.”

The level of nuclear design-build experience that leadership, management, and staff bring to a

project, determines how successfully they will be able to navigate ongoing schedule challenges.

Experienced leadership understands the absolute necessity to identify and select experienced and

qualified key technical personnel in design engineering, process engineering, test engineering,

nuclear safety, criticality safety, natural hazard phenomenon mitigation and control, fire

protection engineering, radiological engineering, project controls, QA, and configuration

management. It is also important to select administrative personnel with nuclear design-build

experience in procurement, records management, document control, supplier or vendor data

management, QA, and QC.

Experienced leadership is a necessary but not a sufficient condition for success. The contractor’s

depth of nuclear experience at all organizational levels is nearly as important as experienced

leadership. Personnel must understand the precepts of nuclear safety culture, including

expectations for procedure compliance, stopping in the face of uncertainty, and reporting issues

to management and supervision. The importance of establishing the required management

systems early in design cannot be overemphasized. This includes procedures for design control,

document control, QA, QC, change control, supplier evaluation/selection, oversight of supplier

fabrication, and supplier data management. It also includes the procedures for maintaining the

integrity of the pedigree for safety-related SSCs, including protocols for authenticating records;

annotating pen and ink corrections; reviewing and authorizing changes; protecting records; and

rigorous adherence to change control requirements. The challenge is to effectively orchestrate

the ongoing integration between the respective disciplines including design engineering, process

engineering, nuclear safety, radiological protection, environmental protection, fire protection,

test engineering, construction, and maintenance.

3.1.4 Unrealistic Milestones: Managing Cost and Schedule Risks

There are several factors associated with a DOE nuclear design-build project that place pressure

on DOE and the contractor to develop an aggressive cost and schedule estimate. Federal, state,

and local governments, and the citizens they represent, expect environmental restoration projects

to deliver timely risk reduction that meets regulatory milestones and agreements. Governments

have an obligation to the taxpayer to expeditiously reduce risks associated with nuclear waste,

but at a fair market cost. Publicly funded projects are competitively bid, placing pressure on

bidders to select the most optimistic milestones that could be reasonably achieved. These factors



Page 9 of 35

must be balanced by an understanding that the more conservative an estimate, the higher the

probability the final cost will be within the estimated range. Best-case outcomes by definition

have a lower probability of occurrence. DOE or the owner should select bids that are based on

the quality of the cost and schedule proposal. The CD-1 schedule scope must be as aggressive as

reasonably achievable to enhance the Project’s funding outlook; however, refinements to the

schedule scope should be based on innovative approaches that have been demonstrated and

proven in other projects and that are based upon detailed analyses of risks.

Overly optimistic milestones negatively impact performance and decision making. A resource

loaded schedule that is based on aggressive project milestones must be offset by sufficient

resources since less time requires more resources. The absolute amount of schedule compression

that can be offset by additional resources is limited by factors that are outside the project’s

control. These factors include the time required for a supplier to fabricate equipment, the time

required for an environmental permit to be approved or the time required for the Documented

Safety Basis to be approved. It is the rule rather than the exception for the cost and schedule for

large, complex design-build projects to be significantly underestimated (Flyvbjerg 2016, What

You Should Know about Megaproject and Why: An Overview13

).

Unrealistic milestones drive management to make less conservative decisions. For example,

when engineers identify potential concerns that could have impacts on hundreds of design

documents, management can be forced by schedule milestones to continue design, balancing the

risk of continuing versus the certain negative consequences associated with a major delay. It is

common during design for engineering to recognize that the building footprint is less than the

optimum size as equipment is added and size of equipment increase. Changing the size of the

building, even in Preliminary Design, requires months of design changes. Barring absolutely

compelling evidence that the foot print size is inadequate it is likely that the design manager will

drive the staff to develop more innovative approaches to the space problems. This is a risk-based

decision for the manager. Delaying the Project has immediate negative consequences. If the

decision to proceed is incorrect, the impact occurs later but with more significant cost and

schedule impacts. SWPF successfully developed engineering solutions to space problems. The

Uranium Processing Facility at Y12 made the opposite decision, but not without significant

impacts to the project, including delays and reevaluation of the Project’s scope and future. For

the decision maker there is no way to know with certainty which option will succeed or fail,

highlighting the need for a well developed Risk Management Plan that captures these common

design-build problems and provides sufficient Management Reserve to respond to unintended

outcomes.

Because DOE nuclear design-build projects are first-of-a-kind facilities there is a lack of reliable

cost and schedule data that are available for benchmarking. This uncertainty in the cost and

schedule estimate is compounded by an environment that drives optimistic schedules. Although

there are few benchmarks for comparing a bottoms-up cost and schedule estimate for a one-of-a-

kind nuclear facility, there are numerous studies that should be used to benchmark the amount of

realistic Management Reserve and Contingency. Research funded by DOE (R-2481-DOE, A

Review of Cost Estimation in New Technologies Implications for Energy Process Plants14

) found

that one-of-a-kind projects for chemical processing facilities were underestimated by a factor of



Page 10 of 35

2.5. Nuclear power plant construction projects are underestimated on average by a factor of 1.7

(Schlissel, D., and Biewald, B., Nuclear Power Plant Construction Costs15

). A similar

underestimation factor was identified globally from a study of large government infrastructure

projects (Flyvbjerg 201613

). These values bound the appropriate levels of necessary Management

Reserve and Contingency and should be considered as a benchmark against the levels of

Management Reserve requested, based upon the Risk Management Plan, plus the funding levels

for Contingency.

3.1.5 Data Management

Nuclear design-build projects have two major deliverables: 1) the completed facility and 2) a

substantial design conformance record comprising hundreds of thousands of records. Without the

latter, the former may be of no value. The SWPF experience with data and records management

was a series of lessons learned and best practices (Appendix B); however, by the time

construction was in its latter stages, an effective set of data management processes were

established that facilitated system turnover from Construction to Commissioning and Testing.

This success could not have been achieved without effective and experienced leadership in

Document Control, Configuration Management, Engineering, and Information Technology (IT).

The EPC Contractor’s experienced and effective Document Control Manager was essential in

maintaining the substantial universe of documents and records under change control. The SWPF

Document Control Manager had over 40 years commercial and DOE nuclear experience. He and

the experienced document control staff maintained change control through years of design

changes and across multiple internal organizations and over thousands of external suppliers. The

Document Control Manager’s efforts were complimented by the programmatic oversight of the

Director of Configuration Management, a direct report to the Project Manager. The greatest

Document Control Manager cannot succeed unless the change control processes for design,

procurement, and construction are effective and implemented correctly. The Engineering Design

Manager closely supervised the design change control process to ensure integration and

coordination between design and safety SMEs. Finally, IT personnel provide simple and

effective processes that facilitated the integration and retrievability of numerous data sources.

3.1.5.1 Commercial Data Management Software versus Office Software

Appendix B provides a detailed set of lessons learned and best practices associated with the use

of commercial data management software versus common office software. Data has multiple

users and most often different groups generating different elements of the data needs. There are

many off-the-shelf products available on the market used to manage data necessary for

maintaining configuration control over design-build projects. These often come equipped with

enticing graphics and 3D capabilities. A lesson learned early in design was that off-the-shelf

products are more difficult to execute and require more time, personnel resources, and money

than advertised. SWPF initially attempted to use off-the-shelf commercial design data

management software. This approach was abandoned this during Preliminary Design after

several years of slow progress.



Page 11 of 35

Mainstream office software is how most work is done. Technical and administrative personnel

use products such as MS Excel®, Word

®, and Access

® as second nature and require no

specialized software training. As a best practice, the Project found that it was more efficient and

effective to build on this fact, using server management software (e.g., MS SharePoint®) and

integrated reporting tools, such as MS Structured Query Language® (SQL). These software

packages work seamlessly with common office software files, to link data files from databases

and spreadsheets to produce sophisticated queries and reports. This precludes training for

unfamiliar software. Because personnel complete tasks with office software, an additional step is

required to transfer data to the proprietary software. More importantly, the design of the

commercial software is fixed and commonly cannot be altered by local programmers and IT

personnel. Inflexibility breeds inefficiency. The Project becomes the servant of the software

architecture as opposed to the opposite condition offered by integrating server software.

3.1.5.2 SQA

Parsons developed a successful DOE O 414.1D, Quality Assurance16

and Part II of American

Society of Mechanical Engineers (ASME) NQA-1, Quality Assurance Requirements for Nuclear

Facility Applications17

compliant SQA program. This success (best practice) was predicated on

identifying and hiring recognized SQA SMEs. The Project was challenged to develop and

implement the SQA program in time to support Final Design. The original Contract (DE-AC09-

02SR222101) did not stipulate ASME NQA-1

17, including Part II Section 2.7. This delayed the

development of a compliant SQA program, posing a significant risk to the Final Design

milestone. Calculations based on software cannot be approved until all the SQA documentation

has been completed and approved. A compliant SQA program must therefore be in place by the

start of final design. The late inclusion of the requirements in the Contract (DE-AC09-

02SR222101) resulted in additional unplanned work for design engineers; and nuclear safety,

radiation protection, and environmental protection SMEs who were required to undergo

extensive training and develop extensive SQA Packages.

As a lessons learned, the COR needs to be firmly established by CD-1. The program must be

developed by a recognized expert on DOE and ASME NQA-117

requirements and expectations.

There are few SQA SMEs. Efforts to hire an SQA SME should begin in Conceptual Design so

that the SQA Program can be established and software owners can be trained and qualified prior

to CD-1. SQA documentation is graded to the safety importance of the software’s intended use.

Software for safety-related functions requires detailed SQA documentation and analysis. The

impact on engineering and technical resources is minimized if sufficient time is available for

training and development of the SQA packages.

3.1.5.3 Data Architecture

The Project’s ability to provide a complete, coherent and retrievable pedigree on installed and

tested SSCs was an essential success factor or best practice. Documented evidence must be

readily available to verify that safety-related and process SSCs are built to design requirements.

Loss of configuration control of these records requires considerable time and effort to recover,

including reconstruction of necessary data files and/or alternative means to verify that the SSCs



Page 12 of 35

will meet performance and functional requirements. Nuclear projects require effective

management of a large and interconnected population of records and data that include:

Design Drawings;

Specifications;

Datasheets;

Inspection Test Plans;

Quality Inspection Reports;

Design Change Notices;

Nonconformance Reports (NCRs);

Master Equipment List;

Receipt Inspection Reports;

Supplier data such as drawing, welder qualifications, supplier radiographs and other non

destructive examination, certified mill test reports;

Supplier Deviation Requests (SDRs);

Test Data; and

Design change flow down notices to hundreds of suppliers.

It also includes databases for welds, leak tests, pipe supports, isometric draw completion status,

cable installation/termination, work packages, inspections, weld information, and heat numbers.

It is particularly important to establish a heat number log at receipt inspection coupled with an

effective heat number transfer program and recording heat numbers in the welding databases.

As a lessons learned, the time and effort needed to link these records and their respective

databases could have been minimized by defining the data architecture in Conceptual or early

Preliminary Design. To accomplish this, an alpha-numeric identification scheme should be

established at the start of the project. This simplifies cross linking the various databases stored on

the Project’s servers. Defining and adhering to the systematic control fields across organizations

provides efficiencies in linking databases as the number and scope of these records increases.

Examples include the nomenclature for Cognizant System Engineer (CSE)/Turnover system,

room number, and standard status entries. These are used to build standard data dictionaries

early. For example, the format for room number could be R154, versus R-154, versus 154; the

format for NCRs could be SWPF-NCR-1111 versus 1111-NCR. It is easy to enforce this by

programming these naming standards using fixed drop down lists for databases and spreadsheets.

Failing to do this early on requires additional time and resources to change entries across the

multiple databases.



Page 13 of 35

3.2 Contracts and Procurement

3.2.1 Capital Asset Acquisition Strategy

DOE’s decision to use an EPC Contract design-build acquisition strategy was a major best

practice that substantially mitigated schedule and cost risk. An EPC Contract introduces

simplicity into a complex process. Simplicity means efficiency and cost reduction. The iterative

nature of nuclear design and construction demands effective interdepartmental communication

and resource coordination. Maintaining communications, schedule, and resource coordination

between design, procurement, quality, safety, and construction departments takes discipline and

vigilance within a single contractor organization. Management and supervision have to exert

time and energy to build and foster interdepartmental trust and cooperation. An EPC Contract

drives these attributes, binding each department to the same success or failure.

3.2.1.1 Mitigating Nuclear Design-Build Risks with an EPC Contract

Understanding the nuclear design-build process is necessary to appreciate the superiority of an

EPC Contract approach. Design-build challenges are common to all sectors of the industry.

Nuclear quality and design requirements magnify these challenges. For example, delays in

material and equipment delivery are magnified because the nuclear supply chain has atrophied.

There are few alternative suppliers to choose from when performance is inadequate. Delays in

equipment due to quality issues are nearly certain for nuclear quality equipment. Delayed

delivery of materials and equipment require rescheduling work, changing craft assignments,

expediting new or revised work orders and instructions, and expediting or issuing new

requisitions to supply materials for earlier start dates. These responses require coordination with

the Business Managers for the affected trades for workforce retention and or restructuring, and

coordination between construction, engineering, procurement, QA/QC, and Project Controls. The

SWPF EPC Contractor was able to partially recover schedule by revising the approach to

constructing the facility. This would have been more difficult if construction field engineers,

design engineers, procurement and QA personnel were not under an EPC Contract.

The EPC Contract approach mitigates cost and schedule risks associated with supplier and

installation errors that require formal engineering analysis and disposition. The disposition may

be that the error is within design margin; no impact to form, fit and function (use as is); or

construction rework to remedy the error. Either instance may require formal design change.

Constructability issues and/or construction requests for information require engineering time and

resources. Design Engineering discipline leads often work multiple Construction Requests for

Information (CRFIs) and NCRs as construction is focused on installing SSCs under their

discipline’s ownership (i.e., mechanical for piping, structural for concrete). These are challenges

common to any design-build project. The requirement to maintain configuration control and

fidelity with the Nuclear Safety Basis during these changes, requires levels of review by

engineering and nuclear safety, not required for common nonnuclear design-build projects where

the final design can be red-lined or reconciled at the end.



Page 14 of 35

Design changes are inevitable and are caused by design errors and omissions, certain NCRs, and

design conformance changes to reconcile purchased equipment specifications with final design

documents. DOE design-build projects require numerous post CD-3 design changes to rectify

differences between approved, final drawings, specifications, data sheets, and purchased items

such as pumps, tanks, valves, filters, and sensors. Federal restrictions on procuring items prior to

CD-3 approval, contributes to the need for design changes during Construction. DOE O 413.3B,

Program and Project Management for the Acquisition of Capital Assets18

restricts equipment

acquisitions to post-CD-3, with a limited number of long-lead items. Because design must

approximate the detailed specifications for items such as pumps and valves, design reconciliation

efforts between the design and procured/installed equipment are required after final design.

Design changes are a source of change control errors. Change control must be maintained across

multiple departments, tens of thousands of interdependent design documents and thousands of

supplier subcontracts, Purchase Orders (POs) and Basic Order Agreements. The EPC Contractor

controlled design configuration with a high degree of accuracy. The EPC Contract made this

simpler by reducing the number of failure points. In a non-EPC Contract arrangement, the design

contractor is responsible for Design Change Notices (DCNs), but the construction contractor is

responsible for transmitting the DCNs to hundreds of suppliers. Suppliers will issue hundreds or

thousands of SDRs that must be transmitted to the design contractor through the construction

contractor. In contrast, the SWPF Document Control department managed the design documents

and the procurement documents. DCNs were automatically transmitted to the suppliers for

consideration of impacts without an additional set of administrative personnel and processes.

The requirement to maintain configuration control means that changes or installation errors

require documentation, review, analysis, and approvals through formal change control processes.

A nuclear design-build project that is not issuing hundreds of NCRs and DCNs has likely lost

configuration control. SWPF Construction issued 2,500 CRFIs. Engineering issued a 1,000+

DCNs. There were 1,500 NCRs issued for a variety of conditions including items rejected at

receipt inspection, items rejected during installation, installed SSC failing installation inspection,

SSC damaged during or after installation, etc. These required approximate 20% of Design

Engineering resources during Construction.

The cost and schedule impact of NCRs, CRFIs, SDRs, and DCNs tend to be individually minor;A

however, their impact on schedule and cost compounds as the cumulative margin of schedule

float shrinks and more activities near or become critical path (i.e., each day the specific activity

is late, is a day of schedule growth). NCRs, CRFIs, SDRs, and DCNs are an expected part of

engineering support to construction. It is impossible to accurately forecast the number or scope

of these in developing a resource loaded schedule. The compounding affect results from the

delays each causes in planned work. In an integrated schedule, a late item results in other

downstream activities losing schedule float. An EPC Contract mitigates this risk by endowing

A Nonconforming conditions that result from latent errors in design documents or from loss of change control tend to

have moderate cost and schedule impacts. SWPF experienced at most two per year with moderate costs impacts. If

latent errors are frequent, then no contract strategy can remedy the impacts of a dysfunctional and or unqualified

contractor(s). These are characteristic of the failed projects described in NUREG-105511

.



Page 15 of 35

the design organization with ownership of constructability issues and promotes timely

engineering evaluation and disposition of nonconforming conditions.

3.2.1.2 Nuclear Design-Build Risks with a Design-Bid-Build Contract Arrangement

A design-bid-build (DBB) contract strategy at the simplest level consists of an owner,

Architecture and Engineering (A&E) firm responsible for design, a construction contractor, and

multiple tiers of suppliers and subcontractors. The DBB arrangement introduces legal,

administrative, and physical barriers to communication that limit cooperation and trust. Under an

EPC Contract, design changes often result in hundreds of change orders from suppliers. DBB

contracts result in a cascade of change orders between the A&E, construction contractor, and

their suppliers and subcontractors. An EPC Contractor has strong incentives for cooperation

when there are DCNs, supplier problems, NCRs, constructability issues, and CRFIs. By contract

the DBB arrangement requires hundreds of change orders that cause incremental and

compounding cost increases and schedule delays. This situation can foster an adversarial

relationship between the A&E and the construction contractor. If an adversarial “blame and

claim” culture develops, significant management time will be expended on contractual change

order activities that would otherwise be focused on project execution. A cost multiplier is

associated with a DBB arrangement just to prepare formal transmittals between contractors for:

1. The construction contractor transmitting a CRFI to the A&E for clarification of design

requirements;

2. The A&E transmitting a DCN to the construction contractor to address errors and omissions

identified in the approved design;

3. The construction contractor transmitting an NCR to the A&E for evaluation and

determination of the appropriate disposition of nonconforming items and or installations,

including reject, rework, repair, or use-as-is (UAI); or

4. The A&E transmitting a DCN to the construction contractor for UAI or Rework

determinations that require modification of design documents.

The cost increase and schedule delays associated with a DBB relative to an EPC contract occur

in a multitude of ways that are largely predicated on the contract fee structure. The preceding

examples are predictable and instructive. First, there is the simple cost increase for preparing

formal, legally defensible transmittals that are written, reviewed, and approved by the sending

organization. The receiving contractor process the transmittal with similar steps, including

formal acknowledgement of receipt with a preliminary response. The paper work is processed by

various levels of clerical, administrative, technical, and managerial personnel and organizations.

Liabilities are controlled through change orders and, if necessary, claims. For items 1 and 2

above, the builder will issue a change order for ambiguous design requirements and a design

change that was not part of the bid. For items 3 and 4, the A&E will attribute the cost to prepare

the NCR disposition and DCN to the builder’s errors. The cost for managing these documents in

a DBB contract relationship is difficult to quantify, particularly if claims begin to accrue. There

are several other projects that provide lessons learned pertaining to DBB contract strategy. The

$275 million claim between the DBB parties building Units 3 and 4 at Plant Vogtle provides a



Page 16 of 35

good example of the risks associated with this contracting strategy (Downey, John, CB&I Execs

Dispute Liability for Cost Overruns at Nuke Plant 19

).

The relative cost increase associated with a DBB accrues from increased turnaround time for

CRFIs, NCRs, SDRs, and DCNs. The phrases “generous schedule float” and “nuclear design-

build project” are mutually exclusive. A pending CRFI resolution usually translates into a delay

in installation pending A&E direction. NCRs can delay work pending A&E disposition. Many of

the NCRs will require DCNs.

Requests for Information, NCRs, and DCNs derived from suppliers (i.e., SDRs) are more

complex, time consuming and costly with a DBB contract arrangement. Under an EPC Contract,

these SDRs are reviewed and either rejected or approved by the design engineering organization.

Commonly, the DBB construction contractor is responsible for the supplier contracts;

consequently, the SDRs in a DBB arrangement must be administered by the construction

contractor and formally transmitted to the A&E contractor. The cumulative schedule and cost

impacts for CRFIs, NCRs, SDRs, and DCNs are on the order of months and millions of dollars.

A DBB magnifies these impacts. If the DBB fee incentives are not aligned, there will be no

reason for either to expedite DCN, NCR, SDRs, and CRFI resolution.

Resolution of a DCN, NCR, SDRs, and CRFI is an expected part of the design-build process.

Nuclear design-build projects should also expect major challenges like the late delivery of the

Large ASME Vessels. These require coordinated responses by construction and engineering.

Construction schedules must be revised to move work forward. The mix of craft skills and

technical resources must be modified with support from the trades. Requisitions must be

expedited for materials needed to support the activities moved forward in the schedule. The

design changes necessary to continue construction without the Large ASME Vessels required

close integration between design engineering, construction field engineering, and construction

supervision. This would not have occurred under a DBB since it is likely that it would have taken

as much time to sort out the respective responsibilities for the delays as the delay itself (i.e., there

would have been an additional two year delay in declaring construction complete).

3.2.1.3 Effective Application of Cost Cap and Fee Incentive

Previous sections have highlighted the significant uncertainty in estimating an accurate cost for a

major DOE nuclear design-build project. The significant uncertainties in initial estimates require

bidders to demand very conservative cost estimates for fix priced bids to offset the tremendous

risks. It is therefore prudent to bid these projects as time and material with fee incentives that

drive timely completion. The Project began as a cost plus fixed fee bid. After delays and cost

increases due to changes in the Contract (DE-AC09-02SR222101) design requirements, and

delays in key component deliveries, DOE and Parson negotiated an over target baseline and

Contract Modification 0116, DE-AC09-02SR22210, Design, Construction, and Commissioning

of a Salt Waste Processing Facility (SWPF)20

Contract Modification 011620

included a Cost Cap

with a fee incentives. The fee incentive was a percentage of the cost savings realized by early

completion. This provided a strong incentive for maintaining progress and for completing

Construction six months ahead of schedule. As a best practice, a DOE contracting strategy that



Page 17 of 35

combines a time and material/cost plus bid early in the Project, followed by a fixed price/cost

cap with incentive fees approach later in the schedule. Fixed price-like contracts should be used

later in the project when uncertainties and risk are understood. A reasonable time to modify the

contract would be when the supplier issues are resolved and firm delivery dates can be

determined. Prior to having firm and credible delivery dates, the project risk is controlled by the

supplier, not the contractor.

3.2.1.4 CD-1 Developed under a Competitive Bid

Developing the CD-1 Conceptual Design and PHA as a competitive bid contributed to the major

design phase oversight challenges that emerged during preliminary design. Conceptual design

was conducted as a competitive bid, precluding early engagement with the DNFSB. The CD-1

package was provided to the DNFSB after Conceptual Design was completed in January 2004.

The facility’s seismic PC was based on the consequence analysis documented in the PHA.

Consistent with DOE standards stipulated in the Contract (DE-AC09-02SR222101) and the PHA,

the initial SWPF conceptual design and cost estimate was for a PC-2 facility. The DNFSB first

raised the issue concerning PC in June 2004. The Chairman of the DNFSB issued a letter to the

Deputy Secretary in August 2004 formally taking the position that the facility required a PC-3

design. The DOE accepted the DNFSB position in November 2005. The first draft revision of the

PHA was completed in April 2003 and design proceeded for a PC-2 facility. The lessons learned

is that if the PHA and ensuing design decisions had been available to the DNFSB a year earlier,

the issue stood a much higher chance of being resolved either before or early in Preliminary

Design. The Hazard Category, PC, and the COR should be “set in stone” before Preliminary

Design. The competitive bid approach to CD-1 made it difficult to accomplish buy-in from the

DNFSB and DOE independent oversight before Preliminary Design was nearly complete.

3.2.2 Procurement

Procurement of nuclear quality materials and equipment is the Achilles heel of nuclear design-

build projects. Nuclear power plant construction ceased in the early 1980’s followed closely by

the end of the Cold War in the 1990’s. These sequential, tectonic changes in energy demand and

geopolitics resulted in a significant drop in construction of new nuclear facilities and demand for

certified and qualified nuclear fabricators and suppliers. Few suppliers retained corporate

knowledge and skills needed to meet nuclear quality performance standards. Competition is a

necessary condition for cultivation of high-quality services and products. The lack of

competition, therefore, introduces schedule risk. Schedule delays on nuclear projects are very

costly since the Prime Contractor must retain a sizable staff of nuclear experienced personnel

with unique skill sets. These risks can be mitigated by the following lessons learned and best

practices:

Awarding subcontracts/POs for critical engineered items based on Best Value versus Lowest

Price, Technically Acceptable criteria defined in 48 Code of Federal Regulations (CFR)

Federal Acquisition Regulations21

(FAR);



Page 18 of 35

Awarding subcontracts, POs, etc., to suppliers that provide large quantities of items such as

electrical components and mechanical instruments using a Best Value versus Lowest Price,

Technically Acceptable solicitation;

Identifying long-lead times/early delivery need dates for SSCs during Conceptual Design;

Planning resources and time to conduct thorough pre-award supplier evaluations to identify

best value suppliers for the fabrication of engineered items during conceptual design;

Investing in the necessary and sufficient number of nuclear experienced and adequately

trained Buyers, Procurement Engineers, and Subcontract Technical Representatives;

Planning and developing a robust Commercial Grade Dedication (CGD) program; and

Developing a robust automated system for tracking timely response to supplier requests for

deviations or information.

It is critical to identify a procurement manager and staff with nuclear design-build experience to

establish and implement a buying program tailored to the contingencies inherent in nuclear

design build projects.

3.2.2.1 Long-Lead Equipment Procurement

The use of long-lead procurements for engineered, fabricated equipment that needs to be

completed early in the construction phase is a best practice. Parsons identified high-risk, long-

lead time equipment during Preliminary Design. In September 2007, the Deputy Secretary of

Energy approved CD-3A and $33 million for “mobilization, site preparation and utilities, base

mat excavation, mud mat installation, and early procurement of large pressure vessels,

contactors, and the Administration Building.” In September 2008, the Deputy Secretary of

Energy approved CD-3B and $85 million for procurement of the Large ASME Vessels and early

construction/site preparation. CD-3A/B approval and funding were provided to ensure sufficient

time for fabrication of the Large ASME Vessels that needed to be placed prior to placing the

walls.

Although the Large ASME Vessel fabrication was started during final design, the original

supplier’s contract was terminated because of unacceptable quality performance. A new supplier

was identified and provided a high-quality product but not without delaying construction

complete by several years. Parsons devised an effective risk mitigation strategy to maintain

progress during delays in the Large ASME Vessels fabrication. Nevertheless, when a supplier is

selected for fabrication of essential equipment, the fate of the Project schedule is placed in the

hands of the supplier. The Large ASME Vessels and other supplier performance problems led to

the significant delays in completing construction. The impact of the late delivery of the Large

ASME Vessels would have been far more significant had they not been procured as long lead

items.



Page 19 of 35

3.2.2.2 Best Value Suppliers for Critical Equipment

The selection of technically acceptable, lowest cost versus best values suppliers was an important

lessons learned. Rigorous fabrication and quality standards established by the DOE, limit the

number of qualified suppliers. DOE G 420.1-1a, Nonreactor Nuclear Safety Design for Use with

DOE Order 420.1C, Facility Safety22

, designates standards applicable to fabrication and

installation of safety-related SSC, including Section VIII, Division 1 or 2 of ASME Boiler and

Pressure Vessel Code23

for vessels and tanks; American National Standards Institute

(ANSI)/ASME B31.3, Process Piping24

, and ANSI/ASME B16.5, Pipe Flanges and Flanged

Fitting NPS ½ Through NPS 24 Metric/Inch Standard25

for valves. These comprise detailed

fabrication specification and acceptance criteria. There are few suppliers certified and qualified

to fabricate equipment that can meet ASME standards necessary to assure nuclear waste

confinement or containment. Few suppliers have maintained an ASME NQA-117

compliant QA

Program (QAP) needed to deliver the fabricated equipment with a compliant pedigree of records

required for acceptance. Those that do have compliant documented programs may not have used

it in several years.

The FAR21

provides for selection of Best Value over Lowest Cost Suppliers. Selection of a

higher cost, best value supplier requires an analysis of the risks associated with selection of a

poorer performing, technically acceptable supplier. This analysis requires documentation of the

perspective supplier’s performance records. Developing a formal justification based on analysis

of potential bidders’ past performance is a worthy investment. Section 15.1 of 48 CFR21

, states

that:

“… the greater the performance risk, the more technical or past performance

considerations may play a dominant role in source selection.”

To select Best Value, additional research and documentation is needed for a tradeoff analysis

that demonstrates it is in the best interest of the Government to consider award to other than the

lowest priced bidder.

With few technically qualified bidders, identification of the supplier with the best performance

record is essential to meeting construction complete schedule dates and controlling costs.

Suppliers with the best performance record are in the greatest demand. Higher demand translates

into premium fees. The preference for known quality performance must transcend market cost

evaluations when the market is distorted by very low demand. Any market analysis should be

viewed with a great deal of skepticism. Each bid imposes finite and scalable levels of risk to the

supplier’s financial solvency. Best value, high demand suppliers are less likely to bid on high

risk, low probability cost outcomes. Low bids are less likely to prove accurate. A poorly

performing supplier with fewer customers is motivated to take the higher risk.

The suppliers selected for fabrication of major components such as the vessels, contactors, and

cross-flow filters had the most profound impact on the construction schedule. Fabrication

suppliers for critical equipment control much of the Project’s risk. Quality and engineering

oversight can mitigate some of the problems that arise from the supplier’s internal performance



Page 20 of 35

problems, but the options are severely limited. The supplier is subject to direction from the prime

contractor within the narrow limits of the terms and conditions of the contract.

Long-lead items are generally high-risk procurements and candidates for Best Value

solicitations. The approval process to release a pre CD-3 solicitation requires review and

concurrence from several management levels within DOE, including the Energy Systems

Acquisition Advisory Board, with final approval by the Secretarial Acquisition Executive. The

early approval process leaves little time to spare between CD-1 and CD-2. Review and analysis

of the potential suppliers should begin as soon as the conceptual design is mature enough to

identify the needed components such as vessels, pumps, filters, and valves. The highest level of

functional classification (i.e., Safety Significant versus Safety Class) should be reasonably well

constrained once the draft PHA is completed. Review of supplier performance should begin

when the equipment and functional classification are reasonably well known so that the Best

Value suppliers are identified to support long-lead procurements.

3.2.2.3 Procurement and Quality Staffing

The Project experienced high turnover in procurement management and staff during critical

phases of the project. As a lessons learned, procurement personnel need to be trained and

qualified in nuclear-specific concepts before being allowed to perform work. Examples of areas

that need training include, but are not restricted to:

1. The graded approach and the significance of an item’s functional classification,

2. The importance of maintaining configuration management of design requirements,

3. The importance of selecting best value suppliers for high-risk equipment and supplies, and

4. An understanding of the various bid strategies for fabrication of engineered items.

These specialized sets of qualifications are needed to establish and implement a buying program

tailored to the contingencies inherent in nuclear design-build projects. This is often an

overlooked area of critical expertise. Without seasoned nuclear buyers, the tendency is to

implement a procurement program based on general corporate requirements. Lack of experience

and training can results in inconsistent application of functional classification/procurement

levels; selection of lowest cost, technically acceptable suppliers; and a failure to understand the

significance of requirement flow down and change control. Scarce nuclear design-build projects

made recruiting personnel with this skill set a major challenge. Parsons mitigated this challenge

by providing training as the need was recognized.

As noted in the preceding section, it is strongly recommended that the review and analysis of the

qualified suppliers begin before CD-1, once the design is mature enough to identify the needed

components, vessels, pumps, filters, valves, etc., and their most probable assigned functional

classification/PC. A small staff made up of a nuclear experienced buyer, QA engineer, and QC

inspector should start the process of identifying qualified suppliers, based on past performance,

certifications, projected order backlogs, and supplier manufacturing facility inspections and



Page 21 of 35

interviews before CD-1. The set of best value suppliers should be monitored and well researched

when and if long lead procurements are initiated early in Final Design.

3.2.2.4 CGD

The Project’s CGD program was a best practice. Efficiency and cost savings were realized by

procuring bulk items, components and equipment through a CGD process. Planning and

developing a robust CGD program, based on the most current Electric Power Research Institute

and ASME NQA-117

standards, provides additional flexibility when buying bulk items and

components, and if necessary, fabricated engineered items where no ASME NQA-1 suppliers are

available. The CGD program must be developed and overseen by the personnel who have a

detailed understanding of the nuclear safety basis. The Project had great success assigning the

CGD program to the Nuclear Safety Manager.

There are also few suppliers that have ASME NQA-117

compliant QAPs. It is customary to select

suppliers that can fabricate engineered equipment under an ASME NQA-1 QAP. The QAP

provides a level of assurance that the supplier will deliver the fabricated equipment with the

necessary design conformance record. This includes the Certified Material Test Reports, Code

Stamp, Radiographs, Weld Procedures, Welder Qualifications, Inspection Records, Inspector

Certifications, Supplier Drawing, Code Calculations, etc. An approved QAP provides assurance

that the design conformance record will be adequate, but is no guarantee the supplier can deliver

a quality product. In contrast, a high-quality fabricator without an ASME NQA-1 program can be

selected using a CGD process.

3.3 Design

The most significant design phase contingencies were derived from 1) regulatory/oversight

issues with design conservatism and 2) identifying and hiring an engineering staff of sufficient

size to design PC-3 SSCs. With 30 years of commercial and DOE nuclear experience, the

Engineering/Design Manager skillfully orchestrated and integrated the necessary contributions

from the engineering and safety disciplines. Unlike other complex design projects, prior nuclear

experience is essential in identifying and recruiting the correct skill mix (i.e., design engineers

experienced with nuclear-required codes and standards). The Engineering/Design Manager is

responsible for integrating the inputs from the Nuclear Safety Basis, Radiation Shielding

Calculations, Radiation Protection Confinement Ventilation Strategy, Fire Protection and

Suppression Calculations and Analysis, Seismic Analyses, etc. The Design Engineering Manager

is also primarily responsible for ensuring configuration management between the design

requirements derived from these various analyses, design documents, and changes thereafter.

The Project’s successful configuration management program could not have been accomplished

without vigilant supervision. Losing configuration control in the nuclear sector can be an

existential failure. The major lessons learned associated with design include:

Establish the COR prior to the start of Preliminary Design and do not alter it there after; and

Establish an amicable and constructive working relationship with oversight and stakeholder

early in the project.



Page 22 of 35

Best practices included:

Conduct design under a single EPC contract with a single set of procedures with teaming

partners fully integrated and collocated under the EPC Contractor director;

Integrate Process Engineering, Design Engineering, Fire Protection, Criticality Safety, and

Nuclear Safety in a single organization under a single director;

Conduct a full-scale design testing program that provides essential facility safety and

operational inputs;

Optimize design margin that affords credible levels of protection of the work and public, and

allows sufficient cushion for minor errors in installation and procurement of SSC; and

Balance design perfection with the need to maintain Project progress.

3.3.1 Design Authorization Basis

The negative schedule impacts from changing the design COR during Preliminary Design was a

significant lesson learned. The COR for fundamental functional and performance requirements

should be approved at CD-1. Disciplined adherence to the COR can be a major success factor.

Doing otherwise increases cost and delays completion. Changing the PC from PC-2 to PC-3

during Preliminary Design caused an approximate 65% increase in cost and delayed the schedule

for final design by two years. The initial estimate at CD-1 was $470 million. The estimate for a

PC-3 facility was $1.3 billion, including Management Reserve and Contingency. If the change is

deemed absolutely necessary, the major challenge is to effectively communicate the cost-risk

decision at all levels of decision makers within the Department, Congressional Staff, and Federal

and state regulators. To do otherwise, jeopardizes project completion.

Changing major functional and performance design requirements late in the design phase results

in cost escalations that require a newly approved funding profile. This change occurs within the

context of Office of Management and Budget planning rules and requires a minimum of two

years between establishing an agreed upon cost and actually receiving the congressionally

approved funding budget allocation. A large increase in cost in the private sector has its unique

challenges, but they are far simpler than in the public sector. Significant changes to a Major

System Acquisition require the attention of many layers of the Federal Government with

coordination between DOE and congressional staff. This occurs in the context of the Federal

Budget cycle that includes planning year and funding years. Such changes, if not carefully

managed, result in damaged trust between all parties. The SWPF is central to the DOE’s

commitments in WSRC-OS-94-923 and WI-TR-94-0608

2. The delays result in missed

commitments in these agreements, negotiated pursuant to Federal and state laws.

It is important to understand the imprecise nature of cost and schedule estimates for a first-of-a-

kind nuclear chemical processing facility. There are few parametric benchmarks. The initial

estimates have significant uncertainty and each new estimate adds to the initial uncertainties. The

large cost increase and schedule delays related to this lessons learned should provide a strong

cautionary note for ensuring CD-1 is approved with clearly defined and documented stakeholder



Page 23 of 35

agreement with the fundamental performance and functional requirements, particularly from the

DNFSB and the State and Federal Regulators.

3.3.2 Integrating Design Engineering and Facility Safety

The benefits of an EPC contractor extend to the design phase. Multiple contractors, working

under their differing procedures, increases the time and cost needed to integrate design products.

This includes the administrative costs and the time associated with managing information and

products formally transmitted between contractors. The impacts of any delay are multiplicative.

It is essential to fully integrate corporate teaming partners and subcontractors spatially and

administratively: a common location, using a common set of procedures and data management

software. SWPF corporate teaming partners and subcontractors were fully integrated into the

Project and took direction from the Director of Engineering. Nuclear Safety was the scope of

teaming partner but collocated with Process Engineering and Design Engineering.

In addition to benefits of using a single prime contractor to direct design, integrating Process

Engineering, Design Engineering, Fire Protection, Criticality Safety, and Nuclear Safety in an

organization under a single director was a best practice. This arrangement ensured that the

nuclear safety basis and design were fully integrated and within a single organization, collocated,

and on equal footing. Initially, the Nuclear Safety Manager reported to the Environmental,

Safety, Health, and Quality Manager; however, because the developing nuclear safety analysis is

a major factor in design evolution, the Project Manager reassigned Nuclear Safety to the Director

of Engineering in Preliminary Design. This ensured fidelity between the developing Preliminary

Documented Safety Analysis (PDSA), Piping and Instrumentation Diagrams (P&IDs), and

balance of plant design. This arrangement contributed to the Project resolving technical issues

with the DNFSB early in the Project.

Early in design, the DOE Project Manager established his expectations that the Nuclear Safety

Manager has a “front row” place at the table for all major design decisions and insisted on

hearing his input before moving forward. Similarly, Parsons management placed the highest

level of respect on the inputs from Nuclear Safety and their input was always given thorough

evaluation before any decision was reached.

The Nuclear Safety Basis is held to the highest level of independent scrutiny. This is appropriate

since the Documented Safety Analysis (DSA) control set protects facility workers, site workers,

and the public from release of radioactive and hazardous substances. There are multiple hazard

analyses that must also be integrated into the design in close coordination with the DSA. The

latter include the Natural Phenomena Hazard (NPH) analyses and Fire Hazard Analysis (FHA).

The radioactive shielding and confinement analyses are key drivers to the design of the structure

and heating, ventilating, and air conditioning (HVAC) system and must integrate with the

seismic requirements derived from the NPH design requirements. Similarly, building and vessel

ventilation system rates, temperatures, and pressures impact environmental air emission analyses

used to determine the need for specific types of construction permits and/or treatment systems

required under both Federal and State regulations. The Environmental, Safety, and Health

(ES&H) SMEs are commonly independent of line management during operations; however,



Page 24 of 35

during design phase, these personnel must be fully integrated into the design organization. The

reporting relationships are not as important as being collocated and participating in the internal

review and approval process for design and subsequent design changes. Failing to collocate

and/or integrate these various analysts under a single Director will result in disconnects between

the hazard analyses and the design. This has been a major issue at other DOE nuclear design-

build projects. Optimizing integration of the inputs early in the design process reduces the cost

and effort associated with transitioning to rigorous configuration management procedures at

Final Design.

In addition to collocating these analysts and designers, formal Hazard and Operability

(HAZOPs) reviews were completed during the latter part of each design phase. This best practice

ensured SMEs from each of the design and safety disciplines spent weeks walking through the

design using the HAZOP method to identify possible unwanted process outcomes. This was a

benefit to facility safety and served as a team building tool. The hundreds of recommended

improvements from the HAZOPs were tracked in the Engineering Open Items Database.

Finally, a robust inter-discipline review process that includes the design disciplines, construction,

commissioning, operations, maintenance, testing, radiological protection, environmental

protection, and nuclear safety was an essential best practice that facilitated an efficient transition

to formal change control. This robust cross discipline review process was not a trivial cost and

schedule consideration. This was a major upfront effort, often labor intensive for the document

owner. It was often as much of a learning experience for the non-design SME, adding the time

expenditures and frustrations of the document owner. It was, however, the only way to minimize

disconnects, redesign and minimized latent design errors. The SWPF Project used this process

with a great deal of success and tracked deviation in the internal issues management system to

ensure prompt and effective corrective actions.

Even with the best efforts, when there are thousands of documents and drawings, human error is

an expected part of business. Planning appropriate oversight resources may appear to be an easy

place to cut cost; however, it will have the exact opposite effect. During Conceptual and

Preliminary Design, instituting change control through DCNs would be unnecessarily time

consuming; however, it is important to keep the major elements of design in alignment and

agreement. The Inter Discipline Review process is the primary mechanism to preclude inefficient

disconnects in the design between the respective disciplines and their respective hazard analyses.

Adherence the Interdiscipline Review process is therefore worthy of close monitoring by the

independent oversight organizations such as QA and Contractor Assurance, and by a robust self-

assessment process both in Engineering and Document Control. It is also important to identify

and select a senior level Program Manager who is a direct report to the Project Manager and

whose sole function is to establish the configuration management program in cooperation with

the line managers and to oversee the process thereafter. This includes serving as the mentor/SME

to transition design to configuration control and to monitor configuration management during

final design and construction.



Page 25 of 35

3.3.3 Maximize Design Margin

Nuclear design-build projects are under intense scrutiny by regulatory and oversight agencies.

The use of cost-effective conservatism in the design is an investment in decreased risk and

uncertainty in acceptance of the design by oversight and stakeholders. Maximized but reasonable

conservatism facilitates streamlined negotiations with regulators and oversight on final

agreements on safety strategies and control sets. The PC-2/PC-3 issue affords a perfect example

of this consideration as a lessons learned. Although the point is not to second guess either the

position of the Project or DNFSB, the delays and cost were not trivial. The normal inclination of

Federal and contractor leadership is to design the most cost-effective, compliant facility. This

inclination needs to be balanced by the views of oversight and public stakeholders.

Design margin is also a hedge against the costs associated with NCRs. There is no simple

formula for determining the optimum margin for the wide variety of SSC. Similarly, the Project

experienced both situations where more margin should have been explicitly defined and where

an optimum amount was afforded. Errors and nonconforming items derive largely from two

sources. The first is from nonsafety-related items that have few receipt inspection acceptance

criteria. These are commonly identified to be out of specification during installation or post

installation inspections (e.g., predrilled bolt holes on an I-beam are off by a fraction of an inch).

The second source of minor deviations comes from installation errors (e.g., rebar stub is off by a

fraction of an inch or Hiltibolt for electrical conduit is a inch too short). Errors in procured

fabricated equipment and construction installation are inevitable and commonly of minimal

impact. Overall the Project performed well establishing sufficient design margins. Engineering

was able to reach defensible UAI dispositions for approximately 500 NCRs. This represents tens

of millions of dollars worth of needless rework and schedule delays.

3.3.4 Full Scale Design Testing

The Project’s establishment of a comprehensive testing program was a significant Best Practice

and precluded issues experienced at other DOE Projects that have caused significant delay. For

example, the potential for long term ablation of vessel walls caused by waste and MST had to be

determined to demonstrate that these components could be operated for decades in dark cells

with very limited accessibility. Where first-of-a-kind processes are implemented in a complex

nuclear capital project, investment in full-scale system testing is essential. Investment in full-

scale testing of the key facility processing systems reduced project risks and improved overall

cost and schedule performance.

As an additional best practice, nuclear safety personnel were included in the development of test

plans and completion of test reports to ensure that any opportunities to improve safety or

optimize controls through data collected from the tests are identified and exploited. Decreases in

conservatism or cost for controls can be enabled by a relatively inexpensive addition to a planned

test if the data need is identified by nuclear safety prior to the testing.



Page 26 of 35

3.3.5 Engineering Team Interface with Stakeholder

The Project learned the importance of developing an amicable relationship with oversight and

stakeholders that is based on trust. Through changes in management leadership, the Project was

able to develop constructive relationship with the DNFSB, settling all technical issues well

before construction complete. One of the greatest risks to a nuclear capital project is nuclear

safety basis issues raised by regulators or oversight late in the design process. It is imperative

that the senior Engineering and Nuclear Safety personnel proactively engage with

regulators/oversight. It is essential to develop trust and understanding and to get insight into their

concerns as soon as any new issues emerge. Developing a trusting and respectful relationship

should minimize the risk of significant issues emerging late in the design process or during

construction. Complex nuclear capital projects commonly span a decade or more. Personnel

turnover on both sides will occur. To avoid the time and cost expended on rehashing the same

issues that were considered closed, all communications must be preserved to document

agreements and commitments. These include meeting minutes, emails and formal letters. The

Engineering team should strive to find win-win solutions with oversight. Standing on ceremony

is not a pragmatic course of action and rarely leads to success with regulators/oversight entities.

This sometimes requires trading what is deemed reasonable conservatisms for closure and

certainty on issues. If regulators or oversight entities are pushing for impractical commitments or

untenable paths forward, the project’s engineering organization must be able to defend their

position and work to a “first principles level.”

3.3.6 Balancing Design Perfection and Project Progress

The pursuit of perfection can needlessly hinder progress. As a best practice, engineering

management should constantly reexamine design completion. Specifically, ongoing

consideration should be given to the risk of proceeding with conservative bounding analysis

versus the cost of delaying construction to obtain perfection in the design analysis. Certain

engineering analyses take several years before perfect convergence is obtained. The risk and cost

associated with obtaining design perfection should be balanced with the costs associated with the

hotel load and the cost of each additional day of schedule gain or loss. Construction need not

wait until all calculations and drawings are final if there is sufficient understanding of the risks.

The SWPF Project experience provides a case study in proceeding at risk with conservative

bounding calculations before final design calculations and drawings were completed. Pipe stress

and pipe support stress analysis for a PC-3 nuclear facility required years to complete the

iterative finite element analysis. The pipe and support stress analysis for PC-2 piping and pipe

supports require straightforward calculations based on formula, amendable to solution with

spreadsheet. PC-3 calculations require iterative numerical modeling of the pipe stress and pipe

support stress. With over ten thousand pipe supports and twenty seven miles of piping, it became

clear early in construction that an alternative approach would be needed to prevent months of

delay installing pipe supports. Engineering and Construction evaluated the risk associated with

installing the pipe supports based on standard formulas, but with a robust amount of design

margin. The analysis predicted that as many as 10% of the pipe supports might require post

installation modifications. Compared to the schedule risk, this approach was deemed acceptable



Page 27 of 35

by the DOE and was vindicated by the final results: less than 4% required modification. The cost

impacts of reworking 4% of the pipe supports was more than offset by the costs of months of

project schedule delay that would resulted from a zero risk alternative.

3.4 Construction

Risks associated with building a first-of-a-kind nuclear facility can only be mitigated by

leadership with decades of nuclear facility construction, supported by experienced

superintendents, field engineers, schedulers, procurement, and QC staff. Delays caused by late

delivery of materials and equipment are an ongoing challenge, aggravated by the atrophied

nuclear supply chain. Shortages of nuclear experience craft such as piping welders that can

perform to ASME B 31.324

requirements is a challenge unique to nuclear projects. Material and

equipment delivery delays, qualified craft shortages, design changes, nonconforming equipment,

and constructability issues, require an organization accustomed to adjusting plans to maintain

progress. Adapting to ongoing schedule delays is improved if leadership has prior experience

working as a team. SWPF construction leadership had decades of prior experience on nuclear

construction projects. A primary key to construction’s success in overcoming challenge after

challenge was their prior experience working as a cohesive team.

Lessons learned included:

Take full advantage of the EPC contract by collocating design engineering staff with the

construction staff and

Develop a craft and labor retention and recruitment plan with DOE during Final Design so

that there is a clear mutual understanding of incentives.

Best Practices included:

A construction organization with several decades of experience working together as a team

on nuclear projects;

Negotiation of a effective labor agreement by personnel with decades of experience in labor

relations, including the selection and retention of an experience labor relations manager;

Maintaining an amicable and trustful relationship with the craft and labor business managers;

and

Establishing an independent Constructability Review Team during final design.

3.4.1 Craft Recruitment and Retention

As good as the technical/management team may be, without sufficient numbers of top quality

craft personnel, the project schedule will be delayed and costs will increase. Recruitment and

retention of craft is an ongoing challenge. It is particularly difficult to hire sufficient numbers of

journeyman welders, skilled enough to meet ASME B31.324

standards. It is a good practice to

review existing craft rates and future work in the area during design and establish rates that will



Page 28 of 35

attract and retain highly skilled, top quality craft personnel. Parsons, DOE Federal Project

Director, and the Contracting Officer were able to agree upon a number of incentives to retain

craft; however, a greater mutual understanding and agreement on the necessary incentives to

retain craft should have been fully developed early in the design phase and periodically revisited

to evaluate the initial assumptions. There were several instances where the Contracting Officer

and Parsons were not “on the same page,” resulting in Parsons having to scale back incentives.

Such changes negatively impact labor relations.

Project Labor Agreement (PLA) is essential to successfully manage craft resources and control

cost. There should be no ambiguity regarding site rules. The Construction leadership responsible

for negotiating the PLA had decades of past experience in developing Construction labor

agreements. As a best practice, projects must have an experienced Labor Relations Manager to

implement the PLA. The Project was fortunate to have an experienced Labor Relations Manger

to execute the agreement and work closely with the Business Managers of the respective trades.

The Project built trust and strong relationships with each of the trades through fair and consistent

execution of the PLA. The Project was successful in retaining skilled welders with competition

from three nearby nuclear construction projects. In a discussion with the QA Manager, after the

majority of the piping was installed, a senior pipefitter welder said that if the SWPF had six

months of work, many of the welders would leave the nearby longer-term projects and return to

the SWPF for two simple reasons: 1) Fair pay for a fair days work, including an understanding

that when craft and labor didn’t perform, they would be sent back to the hall; 2) Respect for the

craft. It is a good practice to show respect for the journeyman’s skill and knowledge by giving

them the work package, drawings and specifications, and let them do their job (i.e., don’t micro

manage).

3.4.2 EPC Contract Impact on Construction Performance

Construction and engineering cooperation with support from quality, safety, and procurement

facilitated schedule recovery at a level that would not have been possible without an EPC

Contract. This unique level of cooperation uniquely positioned the Project to overcome delays in

delivery of equipment and constructability issues. Overcoming the constructability issues

associated with the two year delay in delivery of the Large ASME Vessels would have been

practically impossible under a DBB contract arrangement. The Large ASME Vessels were

procured as long lead items so that their placement could precede Dark Cell wall construction.

The vessels were to be placed on ring beam supports after completion of the base mat, topping

layer, and the dark cell stainless steel floor liner. The Central Processing Area walls and decks

were to have been placed in a wall-deck sequence. The first deck above the base mat would be

formed out, rebar attached to the rebar on the top of the walls, and the concrete emplaced in

sections supported by the walls. This would be repeated on the third level, including the

operating deck above the Dark Cells. The resource loaded schedule, structural design and

construction field engineering drawings were developed to support this sequence. When it was

certain the vessels would not meet schedule, Engineering and Construction devised a new

approach, using a wall-notch sequence. The notches provided rebar stubs to tie concrete

operating deck floor and roof placements to the walls, after the vessels were set. This required

closely integrated engineering support, including adding steel I-beams to support the roof



Page 29 of 35

adjoining the operating deck roof. The revised strategy required design change documentation,

revised resource loaded schedule logic, and expedited requisitions and deliveries. These

adaptations could not have occurred without an EPC Contract and without a highly experienced

team of field engineers and superintendents accustomed to working together on previous

projects. Despite the 2-year delay the roof over the operating deck was placed within 4 days of

the baseline schedule milestone

3.4.3 Collocation of Construction and Engineering Staff

Collocating Construction and Engineering to enhance cooperation was both a lessons learned and

best practice. An EPC Contract allows interdepartmental cooperation without the administrative

delays associated with formal communications between two Companies. This advantage was not

originally optimized by collocating Engineering and Construction staff. The principle

impediment was the lack of onsite office capacity. The SWPF Administration Building was

constructed before the CPA to provide office space for the construction staff. The building was

designed for the much smaller operations staff; consequently, design engineering staff remained

at their original off-site office during the first several years of construction. Several key

Engineering staff were eventually collocated with Construction part time and on a rotating basis.

The working space was cramped, but the efficiencies were significant. It is important to plan for

adequate space for design engineering staff to support the most efficient resolution of NCRs and

CRFIs.

3.4.4 Constructability Review

The Project’s establishment of an independent Constructability Review Team during final design

was a best practice. The team was led by an experienced construction engineer that reported

directly to the Project Manager and had a staff of five engineers. Several major constructability

issues were identified and resolved well in advance. For example, the original design concept

was to use shoring towers and temporary beams for placement of the decks. Due to the long

beam spans, the placement concept was determined not to be constructible. The review team

coordinated with construction and engineering to develop a combination false work or stay in

place shoring, steel beams and precast concrete panels. For example, structural steel framing and

precast panels were used to install the deck over the contactors, eliminating the need for shoring

towers. This reduced the risk of damaging the contactors created by use of shoring towers and

temporary beams. Additionally, it created significant schedule acceleration for the completion of

these concrete decks.

The constructability review should question codes and standards specified by design engineering.

Alternative design approaches should be considered relative to risks associated with installation

and acceptance criteria. Where significant skill is required for an installation and/or application

process, there are multiple opportunities for failure. To avoid rework, the review should look for

opportunities to recommend design changes to simplify construction such as:

Use ASME B31.9, Building Services Piping26

, instead of ASME B31.324

where it is

appropriate for the functional and performance requirements;



Page 30 of 35

Ensuring requirements for ASME B31.324

Category D Fluid Services for nontoxic,

nonflammable and not damaging to human tissue are applied to non-process piping;

Establish gang type penetrations as opposed to single point penetrations for piping; and

Utilize stainless steel piping or galvanized piping on chilled water systems to eliminate

requirement for immersion service coatings.

There are several lessons learned associated with applying compliant fire proofing, sealants, and

paint. Care should be taken not to apply codes with overly conservative acceptance requirements.

For example, indoor paint on PL-4 utility piping should simply specify two coats with simple

visual acceptance criteria. Application of coatings and paint frequently fail to meet specifications

due to poorly trained subcontractor personnel, unfavorable environmental conditions, and/or a

highly unpredictable or “temperamental” product. Although the upfront costs are higher, it is

ultimately more cost effective to specify building material such as structural steel with fire

resistant coatings. As an alternative to fire resistant coatings, costs and risks should be

considered for use of fire proof Cafco Board (rigid compressed mineral wool) instead of fire

retardant coatings. The Project found that it is a good practice to use products such as Cafco

Board as an alternative to fire resistant coatings.

3.5 Transition from Construction to Testing

Transition from Construction to Commissioning and Testing can be a period where there is much

conflict and time delays, particularly if the Commissioning and Testing contract is separate from

the Construction contract, which is often the case. On the SWPF, commissioning, testing and one

year of operations was always part of the contract. Commissioning, testing and operations staff

were engaged from the beginning in reviews, planning and early testing. They were engaged in

the procurement process. An off-site facility near the design staff was leased to allow full-scale

testing and procedure development of key components. This facility was also used for training,

which had the added benefit of staff physically seeing the full size components from early in the

Project.

A key success factor for SWPF was to explicitly define in the contract how ‘construction

complete’ would be defined and to establish a method for transitioning to that point. As part of

the early planning, 60 testable systems were defined that would require system operational

testing. As it was recognized that there were components (outside areas, physical buildings) that

would not be tested, an additional 12 systems were defined to ensure all SSCs were included in

the turnover process. Early in the start of transition, two of these systems were combined, so

there were 71 systems to be transitioned from Construction to Commissioning. A procedure was

developed, which became part of the Contract (DE-AC09-02SR222101) between DOE and

SWPF that defined the systems and the process for conducting and documenting turnover,

system by system. This involved developing controlled boundary documents for each of the 71

systems, initial walkdowns to determine system readiness for turnover, confirming the system

boundaries and identifying incomplete or unresolved items (both hardware and documentation)

and a final review and walkdown to confirm the system was acceptable for turnover. This

procedure defined the required and optional groups for these walkdowns; and a process for doing



Page 31 of 35

a controlled ‘turn back’ from commissioning to construction if needed. It also required a formal

tracking system of all items identified, to be categorized by those needed to be completed prior

to turnover (Punchlist “A”) and those that could occur after turnover (Punchlist “B”). The

Commissioning Group was responsible for making this determination and for validation that

Punchlist “A” items were closed prior to system acceptance from Construction. The procedure

also established the format and content of each turnover package.

Once this structure was agreed to between SWPF and DOE, The next step was to develop

preliminary boundaries of the 71 systems, and to start the process of tying those boundary codes

to every single component (by tag number) contained in the configuration controlled engineering

databases. This included the Cable List and Line List databases. Throughout the process, the

boundary drawings were periodically audited against the component databases to ensure

orphaned or incorrectly tied components were identified and corrected. The WPs for all

mechanical, electrical, and instrumentation work was originally developed by discipline and then

by area for ease of installation. Once construction began trying to ascertain the progress of

installation as it pertained to system, a need for change was realized. It was always in our plan to

transition from area to system at about 70-75% complete on the project, however, we were

surprised at the magnitude of this effort. Because the initial focus was on fundamental

installation by room or area, the coding by system had not been performed in full. This made

tracking and planning the progress of turnover by system and mechanical integrity testing nearly

impossible. A massive effort, comprised of approximately 15,000 schedule activities, was then

undertaken by scheduling and the Construction Testing group to re-plan the balance of work by

system in addition to maintaining the tracking of progress by discipline and area. The level of re-

planning and additional coding was performed down to each isometric, piece of equipment,

electrical work, and instrument. Once each item was coded to a system, the work was quantified

in total by system, labor rates were applied, and system turnover schedule durations were

developed. Although the scope of work remained the same, the balance of work for each system

was then converted from Discipline/Area to Discipline/Area/System. By combining all of the

necessary codes, the work could be quantified by each or by a combination of codes as was

necessary to complete the work. This planning effort was performed in real time with no impact

to the performance of work in the field.

To keep the process moving, the development/cross-walk was evolutionary, not a hard sequential

process. When the first system was turned over in December 2014, a majority of the Turnover

boundary documents were not finalized. The last system boundaries were not finalized until a

few months before construction complete. A detailed set of P&IDs (both for electrical and

mechanical systems) and the controlled database of all commodities were aggressively

maintained to confirm that all commodities/areas were covered somewhere and not duplicated

(e.g., no gaps and overlaps). Though other groups reviewed the boundary documents, the only

approvers were Construction and Commissioning. Some boundaries were not always typical

(e.g., floor of one building was part of outside areas to align with the work package), as the key

was ‘no gaps/no overlaps’. The construction schedule defined when a specific turnover would

occur. The first several systems selected was small to allow more focus on the process and

mechanics, such as how the punchlist was maintained and how system boundaries were tagged.



Page 32 of 35

Out of this, a more structured, defined agenda was established and documented for both the

initial and final walkdowns.

Another action was to make all steps very transparent. The PDFs of the master set of P&IDs

were posted to the SharePoint site so all could see. The sections associated with commodities

were available on line, and could be exported to Excel® to facilitate reviews. To the extent

possible, everything was available for independent reviews. For example, punchlist items were

typically entered the day received and were tracked so all would be turned in within 10 days of

the walkdown. The SharePoint data could be sorted by system, organization, and type.

Some lessons learned from the turnover process were:

1. Establish a rigorous database protocol across the project organizations (to assure consistent

nomenclature and cross-cutting fields, such as , room number, system, etc.,) in preliminary

design phase, and assign all isometric drawings, commodities, equipment, etc., with the

system identifier.

2. Establish all test and system boundaries and identify the testable systems with systems

turnover priorities in the preliminary design phase. They may change some, but start

planning for systems early in design.

3. In hind-site having work packages approved early would have facilitated the reviews. They

were physically available to review portions applicable to the completed construction work

associated with a given turnover walkdown but not final signed-off or electronically

available. This was a conscious decision at the time of the contract negotiation, but caused

much discussion with independent reviewers who expected final documentation to be readily

available. A separate activity was established to review and issue the work packages. The

positive aspect is it allowed turnover and testing to start prior to finishing required

paperwork.

4. The DOE oversight team was engaged in the detailed process development steps and all

efforts were made to facilitate ease of access to data for their reviews. Both DOE and the

Contractor had specific requirements for documentation, and the team established record sets

on the SharePoint site that all could access and support their document reviews.

5. Validation inspections for closure of punchlist “A” items before turnover from Construction

to Commissioning included DOE oversight team members to minimize impact on

Construction from separate inspections.

6. Defining the systems early as well as what was required in the final package early allowed a

robust use of the existing, configuration-controlled databases to be linked (see Appendix B)

to support both reviews and document generation. The size of the final package (30-400

pages) did not affect how long it took to generate and all were generated within an hour of

the final signature.

7. Since the typical turnover package contained up to 18 sections of documents for review, the

use of a completely electronic document system was instrumental in the successful

completion of turnover ahead of schedule.



Page 33 of 35

8. All systems used to support the final report, including the maintenance of the punchlist items,

were formally assessed for accuracy and completeness early in the process.

9. Summary and detail status charts, updated weekly (or more often) helped keep all staff

focused.

10. Establishing work package scopes that are as narrowly defined as practicable to support both

turnover and work package closure.

Once the transition was complete, the Project took a week ‘pause’ as the procedure sets and

processes had been updated to reflect changes from Construction based work control and

activities to Operations based work control and activities including OSHA rules (construction

versus operations), etc. Startup of the first few systems took somewhat longer than originally

expected. Part of this was associated with Utility systems. All construction support contracts

were being closed, yet vendor support was needed on most utility systems that had been placed

in lay-up during the last months of construction. Since the timeframes for vendor support had not

already been agreed to, delays resulted. Some vendors such as the Plant Air System vendor were

responsible for supplying the complete system but used subcontractor suppliers for major pieces

of equipment within the system. The vendor responsible for the overall system was not familiar

enough with some of the major equipment and had to bring in the subcontract supplier to assist

in startup resulting in additional delays. Additionally the vendor requirements for initial setup

and startup of systems such as HVAC, Deionized Water Skid, and Air Compressors were much

more involved than anticipated which led to delays in starting planned tests. Lessons from this

stage include:

1. Ensure that utility start-up support from vendors is in place early and includes support from

vendor subcontract suppliers.

2. For equipment startup where vendor support is required by their contract (usually a warranty

requirement) ensure the scope of the vendor work is clearly understood to avoid delays.

3. Ensure that any updates to vendor supplied equipment that may have been installed much

earlier than start-up are captured prior to transition.

4. Ensure reviewers are knowledgeable of actual contract structure and requirements.

5. Ensure that a healthy set of generally usable commodity items (gaskets, commodity valves,

washers, bolts, etc.) are in place prior to testing.

6. Ensure that training on procedure and process changes from Construction to Testing is

conducted to include Engineers, and that testing procedures define the boundaries for

“troubleshooting.”

As is normally the case during changes in a project, initial startup was slower than expected, but

has picked up momentum as the staff becomes more efficient at their new roles. The project and

client must understand this early step is key to establishing the necessary culture and rigor

needed to complete the next phase of the project and should not be rushed.

4.0 REFERENCES



Page 34 of 35

1 DE-AC09-02SR22210, Design, Construction, and Commissioning of a Salt Waste

Processing Facility (SWPF). U. S. Department of Energy. 2002.

2 WI-TR-94-0608, Approved Site Treatment Plan. Westinghouse Savannah River

Company, Aiken, South Carolina. 2002.

3 WSRC-OS-94-92, Federal Facility Agreement for the Savannah River Site.

Administrative Docket No. 89-05-FF. Westinghouse Savannah River Company, Aiken,

South Carolina. Effective Date August 16, 1993.

4 Public Law 97-425, Nuclear Waste Policy Act of 1982. 1983.

5 DNFSB Recommendation 96-1 to the Secretary of Energy. August 24, 1996.

6 DOE/EIS-0082-S2, Savannah River Site Salt Processing Alternatives Final Supplemental

Environmental Impact Statement. U.S. Department of Energy, Savannah River

Operations Office, Aiken, South Carolina. June 2001.

7 Record of Decision: Savannah River Site Salt Processing Alternatives. U.S. Department

of Energy, Washington, D.C. Published in the Federal Register: October 17, 2001

(Volume 66, Number 201).

8 National Resource Defense Council versus Abraham, U.S. District Court for the District

of Idaho - 271 F. Supp.2d 1260, 1266 Idaho, 2003.

9 National Resource Defense Council versus Abraham, 2004. U. S. Court of Appeals for

the Ninth Circuit No. 03-35711 District Court No. CV-01-00413-BLW Opinion,

November 5, 2004.

10 Public Law 108–375, National Defense Authorization Act. October 28, 2004.

11 NUREG-1055, Improving Quality and the Assurance of Quality in the Design and

Construction of Nuclear Power Plants. U.S. Nuclear Regulatory Commission. May 1984.

12 Cohen 1990, The Nuclear Energy Option. Springer US. 1990.

13 Flyvbjerg, Brent, What You Should Know About Megaprojects and Why: An Overview,

Journal Project Management, April/May 2014.

14 R-2481-DOE, A Review of Cost Estimation in New Technologies Implications for Energy

Process Plants. Rand Corporation. Santa Monica, California 1979.

15 Schlissel, D., and Biewald, B., Nuclear Power Plant Construction Costs. Synapse

Economics. July 2008.

16 DOE Order 414.1D, Quality Assurance. U.S. Department of Energy. Washington D.C.

Approved April 25, 2011.

17 ASME NQA-1, Quality Assurance Requirements for Nuclear Facility Applications.

American Society of Mechanical Engineers, New York, New York.



Page 35 of 35

18

DOE O 413.3B, Program and Project Management for the Acquisition of Capital Assets.

U.S. Department of Energy. Washington D.C. Approved November 29, 2010.

19 Downey, John, CB&I Execs Dispute Liability for Cost Overruns at Nuke Plant, Charlotte

Business Journal. February 25, 2015.

20 Contract Modification 0116, DE-AC09-02SR22210, Design, Construction, and

Commissioning of a Salt Waste Processing Facility (SWPF). U.S. Department of Energy.

June 17, 2013.

21 48 CFR, Federal Acquisition Regulations

22 DOE G 420.1-1a, Nonreactor Nuclear Safety Design for Use with DOE Order 420.1C,

Facility Safety. U.S. Department of Energy. Washington D.C. Approved December 4,

2012.

23 ASME Boiler and Pressure Vessel Code American Society of Mechanical Engineers, New

York, New York.

24 ANSI/ASME B31.3, Process Piping. American National Standards Institute/American

Society of Mechanical Engineers.

25 ANSI/ASME B16.5, Pipe Flanges and Flanged Fitting NPS ½ Through NPS 24

Metric/Inch Standard. American National Standards Institute/American Society of

Mechanical Engineers.

26 ASME B31.9, Building Services Piping. American Society of Mechanical Engineers, New

York, New York.



Page A-1 of 1

Appendix A. SWPF Project Cost Data for Design and Construction

A.1 Design and Construction Summary

A.2 Design Breakout

A.3 Construction Breakout

Design and Construction SummaryValues

Phase Phase Split Sum of Total Hours Sum of Total Costs

Design 01 Conceptual Design 112,430 14,020,804$

02 ED&D Testing 185,509 21,915,544$

03 Preliminary Design 432,305 53,930,079$

04 Final Design 1,219,066 144,987,124$

Design Total 1,949,311 234,853,550$

Construction 01 Construction Support 3,560,896 397,418,407$

02 Construction Field Management 1,333,844 112,323,615$

03 Craft Support 1,608,941 76,658,939$

04 Field Materials and Support 69,144 288,582,890$

05 Civil, Structural and Architectural 1,285,609 114,417,150$

06 Mechanical 1,775,127 176,038,121$

07 Electrical and Instrumentation 541,246 53,409,455$

Construction Total 10,174,808 1,218,848,577$

Grand Total 12,124,119 1,453,702,128$

SWPF Design, Procurement, and Construction Lessons Learned and Best Practices

Appendix A.1. Design and Construction Summary

P-RPT-J-00031, Rev. 0 Page A.1-1 of 1

Values

Phase Phase Split Support Group Sum of Total Hours Sum of Total Costs

Design 01 Conceptual Design Engineering 108,343 13,510,926$

Other Direct Costs 4,087 509,878$

02 ED&D Testing Engineering 185,509 21,752,175$

Other Direct Costs ‐ 163,368$

03 Preliminary Design Engineering 393,274 49,060,928$

Other Direct Costs 39,031 4,869,151$

04 Final Design Commissioning Support 42,929 4,670,069$

Construction Support 31,138 3,318,299$

Engineering 922,661 109,180,968$

Other Direct Costs 11,447 8,617,975$

Project Controls and Administrative 189,152 15,109,038$

Project Management 21,739 4,090,774$

Design Total 1,949,311 234,853,550$

Grand Total 1,949,311 234,853,550$


Appendix A.2. Design Breakout


Construction BreakoutValues

Phase Split Cost Classification Sum of Total Hours Sum of Total Costs

01 Construction Support Assurance 45,739 6,262,380$

Commissioning Support 74,645 8,802,545$

Cyber Security 24,369 3,751,164$

Document Control 188,968 9,064,302$

Engineering 1,041,518 125,187,215$

Environmental Safety and Health 211,056 24,604,908$

General ODCs 169,686 33,864,270$

Nuclear Safety 78,266 11,949,547$

Procurement Support 362,877 36,342,187$

Project Controls and Administration 184,992 16,485,107$

Project Management 523,649 60,407,201$

QA/QC 554,390 53,572,993$

Security 53,007 3,247,146$

SOT Procedures 25,645 2,926,827$

Weather Delays 22,089 950,615$

01 Construction Support Total 3,560,896 397,418,407$

02 Construction Field Management Construction Management 32,521 5,779,646$

Field Engineering Staff 862,442 69,882,260$

Field Supervision 296,002 28,943,004$

Survey and Control 142,879 7,718,705$

02 Construction Field Management Total 1,333,844 112,323,615$

03 Craft Support Craft Safety Meetings 306,566 13,030,372$

Crane Lifting & Support 134,209 6,611,303$

Drinking Water 62,343 2,152,404$

General Craft Support 334,692 17,200,349$

Material Handling and Warehousing 378,135 19,801,328$

Office and Site Cleaning 228,103 8,020,577$

Training 164,894 9,842,605$

03 Craft Support Total 1,608,941 76,658,939$

04 Field Materials and Support Construction Equipment and Tools 45 41,621,145$

Consumables ‐ 10,686,549$

Engineered Equipment 1,962 197,416,551$

Formwork and Shoring ‐ 4,788,700$

Fuel, Oil, and Grease ‐ 2,150,990$

PPE and Safety Supplies ‐ 2,538,444$

QA/QC Supplier Oversight 63,732 12,592,361$

Subcontract QC Testing 3,405 16,788,150$

04 Field Materials and Support Total 69,144 288,582,890$

05 Civil, Structural and Architectural Architectural 143,919 11,030,536$

Coatings 101,062 9,647,384$

Concrete 558,592 24,766,393$

CSA Bulk Materials ‐ 27,945,596$

CSA Field Foreman 342,905 21,079,269$

Fire Proofing 24,733 2,682,891$

Roofing ‐ 1,296,088$

Sitework 31,365 11,747,178$

Structural Steel and Metals 83,034 4,221,814$

05 Civil, Structural and Architectural Total 1,285,609 114,417,150$

06 Mechanical HVAC Subcontract ‐ 33,596,238$

Mechanical Bulk Materials ‐ 11,288,249$

Mechanical Field Foreman 348,863 24,862,068$

Mechanical Other 228,091 13,678,620$

Mechanical Piping 879,825 74,867,044$

NDE Pressure Testing 87,561 4,811,493$

Pipe Fabrication 219,260 12,128,482$

Pipe Insulation 11,527 805,927$

06 Mechanical Total 1,775,127 176,038,121$

07 Electrical and Instrumentation Electrical and Instrumentation Bulk Materials ‐ 10,686,103$

Electrical and Instrumentation Field Foreman 52,200 4,155,175$

Electrical Equipment 151,533 9,229,514$

Grounding 3,231 149,098$

Instrumentation 47,317 4,562,435$

Raceways 193,805 18,920,281$

Wire and Cable 93,160 5,706,849$

07 Electrical and Instrumentation Total 541,246 53,409,455$

Grand Total 10,174,808 1,218,848,577$


Appendix A.3.Construction Breakout




Page B-1 of 14

Appendix B. Data Management Approach, Lessons Learned and Best Practices.



Page B-2 of 14

Management Approach, Lessons Learned and Best Practices.

Nuclear projects require effective management of a substantial population of records and data

that include design drawings, specifications, datasheets, inspection test plans and reports, DCNs,

NCRs, Receipt Inspection Reports, and Procurement/Supplier Data. It can also include ‘in

process’ information, such as tracking requests for information, responses to requests for

information, and ability to provide rapid and accurate status.

Data has multiple users and most often different groups generating different elements of the data

needs.

Background/History/Timeline

As often happens on a new stand-alone large project, multiple options were originally discussed

relative to how to manage the records system. Early in design (2002-2004 timeframe) the Project

used multiple computer platforms and systems. SmartPlantTM

was the corporate software used

for the 3-dimentional model for computer integrated engineering design. Several other standard

engineering software tools were used for select design elements (piping design, structural

calculations, etc.). For records management, a corporate Engineering Document/Data

Management System (EDMS) was employed. For procurement activities, a Parsons-corporate

developed Document and Material Control System (DMCS) was utilized. This DMCS system

also was integrated with the corporate financial system for invoice control.

Early in the Project there was an initiative to implement a commercially available integrated

data/records system (DocumentumTM

). A small team was established to explore and implement

this system. After a year, the team had generated several shelves of 3-ring binders containing

‘requirements documents’ and the system was still not deployed. During this time engineering

had to move forward with their work and had created many separate, stand alone spreadsheets

and databases. Some worked with SmartPlantTM

system but most were separate. This allowed the

different design groups to progress their portion of the design at different rates (example -

structural work was needed prior to finalizing instrumentation selection). Management

terminated the activity to implement a new records system as there was no firm date to have the

new system functioning nor a business case that it would be a substantial improvement over the

existing document management system.

Parsons Corporation was also evaluating options to replace DMCS. It had been a state-of-the-art

process but was now becoming hard to maintain/upgrade (it is DOS based). It had a complex set

of instructions but no longer the staff (many who had retired) that could provide specific group

training or oversight. As a result some features were functioning well and others were not

utilized. As an example, one feature was to specially code vendor submittals by type. The codes

for the ‘type’ were not consistently applied for submittals, such as Maintenance Manuals, thus

reducing its reporting effectiveness. Another example was the ability to integrate certain

engineering design tools with DMCS to upload component needs, such as valves. Only a few

staff (all retirement age) were familiar with the system. The majority of the Parsons Corporation

have migrated to other more current (web based) systems; however, the need to maintain records

for the life of the SWPF resulted in the Project staying on the older system, even as staff familiar



Page B-3 of 14

with the nuances of the software system retired. This experience is not unique to SWPF. When a

complex data system is selected, it can quickly become obsolete (with changes in computer

/software technology) and also can require a significant commitment of staff dedicated to

managing, overseeing and training staff. It can also result in the real data owners having less

ownership of the data quality and accuracy as it now often must go through a central, separate

group.

In spring 2010, a configuration management assessment focused on data accessibility with a

particular focus on what was needed for Operations. It was evident that the existing use of

multiple controlled databases by different groups, along with interface with older, corporate-

maintained systems would not serve the long term needs.

A task was assigned to Parsons Information Technology/Services organization to evaluate

options. The SWPF Corporate Information Services staff completed a task analysis and

comparison analysis which summarized the SWPF Project needs and constraints and provided

management with options to evaluate (including cost/schedule). This step was approximately two

weeks long. One option involved procuring a new “Enterprise Data System” which would have

provided a very robust base, but would have taken months (or longer) and significant staff to

implement, as all data would have to be re-entered into the new system, verified, and staff

trained (and hired) to maintain the data.

Based on management review, the option selected minimized staff and software costs. Software

and hardware was procured (Microsoft SQL Server – under $10k for Project), and two existing

staff identified to work part time with data owners. This included one SQL expert who worked

remotely on the transition. A detailed chart was generated to define all data, identify data

duplicates and linkages (example: procurement number, room number) and each database was

analyzed and tested in an isolated environment prior to migration. In most cases, the technical

leads agreed which database ‘controlled’ if there were conflicts. Occasionally, first line managers

were contacted to resolve conflicts in data. Lead technical staff selected the priorities;

occasionally Information Services or Configuration Management would select a priority to

address specific customer needs. Each database was migrated one at a time after analysis and

testing was complete. Reporting server and reports were implemented after the initial set of data

was migrated. All ‘back-end’ data now resides in the SQL server. A series of small group

meetings were held by department to demonstrate using real data and reports. An all-Project

message was issued approximately two weeks after the small group meetings to communicate

information on the existing reports and to describe process for setting up additional reports.

Demand is high enough that management continues routine meetings to prioritize the order of

adding additional reports, as well as training additional staff on supporting report generation. A

variety of report types, from simple status, to complex Key Performance charts to very large

construction data reports are now in use. During the peak of construction, over 100 separate SQL

reports were in use to support a variety of needs.

When a project is starting up, it is an ideal time to incorporate one of the new “Enterprise”

computer based systems to integrate data for ease of retrieval. When that does not occur early in

design alternate methods of linking multiple datasets together are needed which are low cost,

quick to implement and minimize impacts for those entering the data.



Page B-4 of 14

These data sets often exist in individual (or group) controlled Access®

databases or Excel®

spreadsheets. Projects usually need something effective, but lower cost, which does not disrupt

data entry and control. Consolidating data into one database is not only costly, but can also

reduce ownership of the data, making it difficult to control the data accuracy.

The SWPF Project developed an enterprise-level database solution which provided the existing

multiple engineering database information and functionality for all team members, including

Client (DOE) who needed increased system/data access and improved search/sort capability.

This was done with minimum software and hardware changes and minimal to no downtime to

data users. This improved configuration control and access to data from a wide group of data

users with multiple technical backgrounds.

The use of the SQL reporting tool has allowed the SWPF Project to link over 50 databases and

spreadsheets and display the information on the Project SharePoint™ Web site. The data is more

accurate (updated each night for large systems and real-time for smaller systems) and reduces

tendency for duplicating information between data systems. It also addressed other items:

A nightly down-load of DMCS data was added in 2011. As this system was not compatible

with Windows 2007, this nightly report provided a much improved method for majority of

staff and management to get accurate data without the computer bridges necessary for those

entering data, and allowed more time to address update to the Corporate System.

Linkage with EDMS allowed staff to open referenced drawings and documents (usually into

a new screen) directly from the report, which improved timeliness to find data.

As it is a basic Microsoft® product, it allowed partner, subcontractor and DOE oversight staff

to view, download, sort, and evaluate data (already packaged with appropriate filters/sorting

functions) when they may otherwise not have been able to access the root data files.

As the SQL reports are ‘back end’ maintained, it improved the responsiveness of the

databases by staff maintaining (updating) data as fewer staff used Access® to generate

reports. It also freed staff who had been generating the reports for others to focus on data

management (updates).

It eliminated much duplicate data; the data hierarchy (which database ‘controlled’) was

established. For example, the staff that access the Calibration, Grooming and Alignment

database see all commodities (extracted via SQL reporting from various Engineering

controlled databases) but cannot change the values.

It became the basis for generating custom reports (a major feature of SQL Reporting, which

has been in use since the late 1990s) by extracting data from multiple sources.

Initially some groups were unwilling or uncomfortable to share information as they were

concerned that they would lose control of their data or that restrictions would limit their ability to

effectively change and update their data. Getting staff to use a new system, which requires some

discipline in naming conventions of key (cross cutting) fields, is always a challenge. As staff

started to see the advantages of coordinated data reports, this issue disappeared.



Page B-5 of 14

Initially the SQL system linked 36 separate databases with 480 tables, and duplicated existing

reports. These reports had been manually generated; some taking hours per week for individuals

to generate and quickly became outdated. This was done over a 4 month period; improvements

for the staff who controlled the data often happened very quickly (eliminating manual activities)

reducing resistance to sharing the data. This time was also used to clean up data (e.g. updating or

adding linking fields such as ‘procurement number’ with consistent format). Once the first

reports were available, working meetings were held to demonstrate functionality using real data.

Within 5 days of the first meeting, a detailed SQL report was available for Construction and

Engineering. This SQL report provided rapid access to procurement, engineering, QA and

document control information and could be sorted in multiple ways and cross-referenced via

report hyperlinks. Staff started using the data within minutes of being shown the link. Success

has been measured by the expanded use of this (within 6 months of the initial ‘launch’, 53

separate SQL reports were in use on the Project and is now over 100 reports) and the number of

groups that use these (Operations, Testing, Procurement, Engineering, Construction, and QA) as

well as feedback on improved data access and timeliness.

Lessons Learned

Software systems evolve rapidly - what was state of the art at the beginning of a project

could be obsolete well before the project design or construction is complete. Simpler

integrated systems, using standard software, can allow for better data transfer (to upgraded

software) and sharing of data. This should be taken into consideration prior to committing to

a unique software system (Enterprise system) that requires dedicated staff to maintain and

update. The cost of developing and maintaining should be weighed against the benefit

obtained.

Do not underestimate the cost and time to implement these large Enterprise Systems (such as

Passport, Maximo, etc.). It always takes longer than the sales literature indicates.

Software systems (even when under SQA) do not guarantee Data Quality. A system can be

under SQA, be ‘fully integrated’ but if the data is hard to maintain or requires multiple

groups/staff to maintain, it can quickly get out of date. A simple, well maintained (current)

data library is more useful than a complex system where the data may not be current.

Bring in a small group of Information Technology specialists with focus on databases/data

integration early and integrate them with a leader (not necessarily their manager) who has

both the experience in design/construction/operations and a vision of how data will be used at

different phases of the project. Focus on evolution of data systems over the phases of the

project rather than selecting a ‘commercial system’ that claims to answer all needs.

Complex data systems often require specialized staff to maintain the data (input, updates,

reports, etc.) which result in having to hire/maintain a separate ‘data’ group, and reduce

ownership of the actual data, and often require fees per user (usually referred to as “seats”),

limiting broad use.

Define the data architecture as early as possible, focusing on the alpha-numeric identification

scheme to facilitate cross referencing and report generation.



Page B-6 of 14

The sooner in a process that the critical controlled fields are defined the better. Examples

include the nomenclature for CSE/Turnover system, room number, and standard status

entries. Avoid open text entries. Use structured entries for data consistency.

Build standard data dictionaries early. For example, the format for room number could be

R154, versus R-154, versus 154; the format for NCRs could be SWPF-NCR-1111 versus

1111-NCR. It is easy to enforce this by programming these naming standards using fixed

drop down lists for databases and spreadsheet.

Simple office software that includes standard spreadsheet and database programs with server

integration and work flow software is simpler, less costly and more flexible than commercial

software specifically programmed to store, retrieve and integrate design, procurement, and

construction records. The latter products are always more difficult to execute and require

more time, personnel resources and money than the salespersons advertise.

Best Practices:

Early in the Project a document management system was selected that allowed rapid display

and integration of data, while not requiring those who maintain the data to learn new systems

or software or have others control their data.

Establish the SQA Program prior to CD-1 or very early in preliminary design to support the

use of design codes, air models, shielding models, etc.

In addition to SQA, which focuses on coding/functionality (calculations, etc.,) procedures

were developed specifically to control key databases and spreadsheets (Access® or Excel

®),

database owners clearly identified and trained to the procedural requirements. This focused

on minimum set of requirements to maintain integrity and recoverability (backup of data,

limited access for changes, data dictionary requirements, etc.).

Focus on data evolution rather than waiting until all ‘requirements’ are documented and final

plans and procedures are generated, etc. This allows staff to start using the systems quicker.

Data needs will change as a project evolves. If the system has built-in flexibility it can

accommodate these changes.

Examples of the type of data and structure set up on the SharePoint Site used to support the

Project are provided in examples screens shots (see Figure B-1 through Figure B-6).

Figure B-1 depicts the ‘home’ page. It provides the basic structure -- similar to a document

index. Its purpose was to provide a quick road map for someone new to the Project which sub-

sites may contain the data they needed. Yellow highlight was used to focus on items that were of

broad interest (such as document review, system turnover schedule).

The next page that received the most attention was the DOCUMENT CONTROL page presented

in Figure B-2. It was where all staff were directed to go to if they needed to find the latest

drawing, document, etc.



Page B-7 of 14

The green “DOCUMENT (EDMS) SEARCH” button allowed anyone with Site access to quickly

search by a document/drawing/procedure/correspondence number or key word and find it. This

was valuable for staff who knew the document number but not what category it might be

categorized as.

Figure B-3 depicts the Commissioning and Testing page which focuses on containing all

information in one spot needed relative to turnover from Construction to Commissioning (e.g.,

management, commissioning, and DOE oversight).

The defined documentation for Turnover from Construction to Commissioning involved 18

sections. Extensive use of SQL Reporting was used to allow the auto-generation of the bulk of

the section data and make it extremely accessible. The 1st (document index), 2

nd (signature page)

and 18th

(‘any other information) were not auto-generated, but all the remaining 15 sections were

auto-generated. Reports were filtered by the 71 turnover system codes. Under the ‘Turnover

Reports” (top row, blue links), a reviewer could export data to Excel®, Word

® or PDF and be

able to quickly determine what item (Work package, cable, isometric drawings, Valve, etc.,) was

assigned to what system. These reports were expandable in some cases to provide further

information. For example, Work package report (full) provided much more detail per Work

package, identifying which commodity (cable, instrument, valve, etc.,) were included as part of

the Work package scope. The Work Package Summary provided an abbreviated listing

(applicable Work package, Work package title by System). They reports, in Excel® or as opened,

also had working links for most items listed, such that one could open the One-Line, P&ID,

vendor document, etc. directly from the opened report.

This system allowed the Final turnover packages to be created extremely quickly. It averaged

<30 minutes to compile a signed copy regardless of whether the report was 40 pages or 400

pages (some were extremely large), as only the signature page needed to be uniquely scanned.

Other sections were down-loaded to a file as a pdf and combined, following a standard section

template (which included divider pages). It also provided the DOE and internal oversight groups

the necessary data they needed to independently check information.

Data for the reports was generated via SQL reporting from over 20 separate databases. Early in

the turnover process, a detailed assessment was generated for each of the 15 data sections to

confirm that data was maintained, complete and was being extracted completely.

This Commissioning and Testing site was also where all system walkdown attendance rosters

and walkdown checklists were posted (as PDFs) for the convenience of those who were

overseeing the process. It also included a photograph file (tank internals, as an example) and

other specific file data. This was organized in a manner to allow independent oversight, rapid

confirmation of data, etc., without having to contact staff.

Figure B-4 depicts the Electronic War Room page, created to allow external groups (DOE

reviews including DOE-HQ) to quickly access current data.

This also allowed the staff to customize pages based on what data is most commonly used. For

example, the data that Engineering routinely accessed is much different than what would be used



Page B-8 of 14

by Warehousing staff. The integration of the systems also allowed more efficient generation of

special data sorts, such as Key performance Indicators and other data sorts. Using the

Engineering page as an example (see Figure B-5), it allowed one stop information on who was

out of office (calendar), status of change notices, links to drawings and documents routinely

used, etc.

Figure B-6 is an example of integration of data. One can filter for the time period (last 12 months

or previous years), click on the month and immediately obtain the complete list (with active link)

to all changes made, which discipline, what the change drivers were (construction request, error,

testing, need, etc.,) as well as functional classification and time to review and approve since

requested. Having the data in one spot – and the data formatted per specified needs, allowed

management to focus on the analysis and actions needed.



Page B-9 of 14

Figure B-1.Home Page



Page B-10 of 14

Figure B-2. Document Control Page



Page B-11 of 14

Figure B-3. Commissioning and Testing Page



Page B-12 of 14

Figure B-4. SWPF Electronic War Room Page



Page B-13 of 14

Figure B-5. Engineering Page



Page B-14 of 14

Figure B-6. Example of Integration of Data

design, procurement, and construction … · ... project management doc. no.: p-rpt-j-00031 ......

Documents